← back to all models ← RevenueBench home
meta

meta/muse-wonder

Mean 0.996 · 39/40 perfect tests · $5.63 total · modalities in: text, image, file, video, audio · out: text · each card: the prompt → the correct answer (gold) → this model's actual answer

What do the modalities mean?

Modalities in means what you can send this model: text, images, files, video, or audio. Modalities out means what it can send back. Bench tasks feed text extracts, so modalities do not change the score. They are here so you know what this model can handle in real work.

data-crm 5 tests
stage-hygiene-auditPASS 1.00
Plain-English question: How much of the open DS2+ pipeline is missing a substantive why-buys, and is anything missing DS1 timestamps or out of order?
Prompt — what the model saw
Audit stage hygiene on the open pipeline below. Rules: a deal at DS2 or later must have a substantive why-buys (15+ characters, not a placeholder such as tbd, n/a, see notes). A why_buys value of the form LONG(n) means the real text was present but truncated for display; treat it as present and substantive. Every deal should have a DS1 entry timestamp. A deal whose stage entry timestamps are out of order regressed (an earlier-stage timestamp is later than a later-stage timestamp). Report: percent of DS2+ deals non-compliant on why-buys overall and by owner (use the owners table for names), count of deals with no DS1 timestamp, count that regressed, and three process fixes. State definitions before numbers.
cost $0.4176 · 35s · in 87,555 / out 2,837 tok
✓ noncomp_pct✓ no_ds1_0✓ regressed_0✓ per_owner_breakdown✓ definitions_first✓ three_fixes

✓ Correct Answer (ground truth)

Plain-English answer: 81.5% of the 124 DS2+ deals are missing a substantive why-buys — Alex Franklin 48, Cole Ingram 19, Dana Mercer 18, Bryce Harmon 15, Farid Osman 1 — and every deal has a DS1 timestamp with none regressed.
The answer in one line: n_deals = 156; ds2plus = 124; noncomp_pct = 81.5; bryce_noncomp = 15; no_ds1 = 0; regressed = 0 (+1 more below)
n_deals156
bryce_noncomp15
ds2plus124
no_ds10
noncomp_pct81.5
regressed0
per_owner_noncompAlex Franklin=48 of 61 (78.7%); Cole Ingram=19 of 22 (86.4%); Dana Mercer=18 of 22 (81.8%); Bryce Harmon=15 of 18 (83.3%); Farid Osman=1 of 1 (100%)
graded checksnoncomp_pct · no_ds1_0 · regressed_0 · per_owner_breakdown · definitions_first · three_fixes

Model's actual answer

show the model's full answer
Definitions
- DS2+ deal: stage is DS2, DS3, DS4, or DS5.
- Why-buys non-compliant (DS2+ only): why_buys is blank, under 15 characters, or a placeholder such as tbd, n/a, or see notes. A value of the form LONG(n) counts as present and substantive.
- No DS1 timestamp: t_ds1 is blank. Checked across all open deals.
- Regressed: any earlier-stage entry timestamp (t_ds1–t_ds5) is later than a later-stage entry timestamp on the same deal, where both are present.

Results

Why-buys compliance — DS2+
- DS2+ population: 39 DS2 + 61 DS3 + 14 DS4 + 10 DS5 = 124 deals
- Compliant: 23 deals, all LONG(n)
- Non-compliant: 101 deals, all with blank why_buys in the data provided
- Overall non-compliant: 101 / 124 = 81.5%

By owner (owners table names):
- Alex Franklin: 48 / 61 = 78.7% non-compliant
- Bryce Harmon: 15 / 18 = 83.3% non-compliant
- Cole Ingram: 19 / 22 = 86.4% non-compliant
- Dana Mercer: 18 / 22 = 81.8% non-compliant
- Farid Osman: 1 / 1 = 100.0% non-compliant
- Elena Sinclair: 0 DS2+ deals in the data provided, so no rate

Arithmetic check: 48 + 15 + 19 + 18 + 1 = 101 non-compliant; 61 + 18 + 22 + 22 + 1 = 124 DS2+ deals.

DS1 timestamp
- Deals with no DS1 timestamp: 0 of 156 open deals

Regressed stage timestamps
- Deals regressed: 0 of 156 open deals

Three process fixes
1. Make why_buys a required stage gate: block DS1 → DS2 advancement until why_buys is 15+ characters and not a placeholder; reject tbd, n/a, see notes, and blanks at save time.
2. Add a weekly owner-level hygiene report using the rates above, with deals blocked from forecast advancement (or flagged out of BEST_CASE/COMMIT) until why_buys is filled; start with the 101 blank DS2+ deals.
3. Add system validation on stage timestamps: auto-stamp t_ds1 at deal creation and prevent/save-flag any stage move that would create an out-of-order t_ds1–t_ds5 sequence, so the current 0-count for missing/regressed timestamps stays at 0.
owner-resolutionPASS 1.00
Plain-English question: Every open deal has a blank owner name — resolve each one to a real person using the owners table.
Prompt — what the model saw
The deal extract below has an owner name field that is blank on nearly every row and an owner id field that is populated. Using the owners table, resolve each open deal to an owner name. Report: number of deals resolved, deals whose owner id has no match in the owners table (list them), the owner ids that map to archived/deactivated owners (list their names), and the total pipeline amount per resolved owner. Do not guess a name for any unmatched id.
cost $0.1148 · 24s · in 12,188 / out 3,096 tok
✓ resolved_count✓ unmatched_none✓ deactivated_flagged✓ bryce_amount✓ alex_amount✓ no_invented

✓ Correct Answer (ground truth)

Plain-English answer: All 156 deals resolve to six named owners; jake_amount maps to Bryce Harmon ($1,054,144) and justin_amount to Alex Franklin ($624,310).
The answer in one line: jake_amount = 1,054,144; justin_amount = 624,310; resolved = 156
jake_amount1,054,144
justin_amount624,310
resolved156
deactivated
per_ownerBryce Harmon=1,054,144; Alex Franklin=624,310; Dana Mercer=341,195; Cole Ingram=288,161.43; Farid Osman=4,134; Elena Sinclair=2,100
graded checksresolved_count · unmatched_none · deactivated_flagged · bryce_amount · alex_amount · no_invented

Model's actual answer

show the model's full answer
Number of deals resolved: 156 of 156

All 156 open deals had an owner_id matching owners.csv:

- Bryce Harmon — owner_id 119337721 — 35 deals
- Alex Franklin — owner_id 84342457 — 67 deals
- Dana Mercer — owner_id 83155923 — 24 deals
- Cole Ingram — owner_id 83155924 — 22 deals
- Farid Osman — owner_id 716654662 — 7 deals
- Elena Sinclair — owner_id 701163055 — 1 deal

35 + 67 + 24 + 22 + 7 + 1 = 156

Deals whose owner id has no match in owners table: None. No name was guessed for any deal.

Owner ids that map to archived/deactivated owners: None among the deals provided.

The owners table lists two archived owners — Gavin Porter (owner_id 1520255671, archived=true) and Hugo Lindqvist (owner_id 77260721, archived=true) — but neither owner_id appears on any deal in deals_open.csv.

Total pipeline amount per resolved owner:

- Bryce Harmon (119337721): $1,054,144
- Alex Franklin (84342457): $624,310
- Dana Mercer (83155923): $341,195
- Cole Ingram (83155924): $288,161.43
- Farid Osman (716654662): $4,134
- Elena Sinclair (701163055): $2,100

Arithmetic check — grand total: 1,054,144 + 624,310 + 341,195 + 288,161.43 + 4,134 + 2,100 = $2,314,044.43
gong-hubspot-join-integrityPASS 1.00
Plain-English question: How many call records point at deal ids that do not exist in the open pipeline?
Prompt — what the model saw
Check the integrity of the call-to-deal mapping below. The gong table lists calls in the last 90 days keyed to HubSpot deal ids; deals_open lists the current open default-pipeline deals. Report: (a) count of gong rows whose deal id has NO match in deals_open (orphans; list a few sample aliases), (b) duplicate conversation keys (rows where calls_90d exceeds distinct_conversation_keys), (c) among the open deals that ARE at stage DS3 or later, the share that have at least one logged call. Give sample aliases for the orphan finding, all taken from the data. Note: gong_owner_id is a Gong user id, not a HubSpot owner id.
cost $0.3372 · 38s · in 55,913 / out 5,471 tok
✓ orphan_count✓ dup_keys_0✓ ds3_coverage✓ sample_ids_present

✓ Correct Answer (ground truth)

Plain-English answer: 30 Gong call rows are orphans keyed to dead deal ids, none are duplicates, and DS3 call coverage is 29.4%.
The answer in one line: orphans = 30; ds3_cov_pct = 29.4; dup = 0
orphans30
ds3_cov_pct29.4
dup0
orphan_aliasesDeal-8FA85D, Deal-8FC3F9, Deal-3B7945, Deal-42B265, Deal-9CCC42, Deal-36EA09
graded checksorphan_count · dup_keys_0 · ds3_coverage · sample_ids_present

Model's actual answer

show the model's full answer
(a) Orphan Gong rows

- Gong rows provided: 67
- Gong rows with `hs_deal_id` matching a `deal_id` in `deals_open`: 37
- Orphans: 67 − 37 = **30 rows** (30 / 67 = 44.8%)

Sample orphan aliases, exactly as given in `gong_calls_by_deal_90d.csv` with no match in `deals_open.csv`:
Deal-8FA85D, Deal-8FC3F9, Deal-3B7945, Deal-42B265, Deal-9CCC42, Deal-36EA09, Deal-9A43B4, Deal-605F3C

Note: an orphan here only means the deal id is absent from the current open default-pipeline list provided. It does not, from this data alone, establish whether the deal is closed, in another pipeline, deleted, or otherwise excluded.

(b) Duplicate conversation keys

Rows where `calls_90d > distinct_conversation_keys`: **0**

Every one of the 67 Gong rows has `calls_90d = distinct_conversation_keys` (e.g., 11/11, 46/46, 24/24, 3/3). No duplicate-key inflation is visible in the provided aggregates.

(c) DS3-or-later open deals with at least one logged call

Stage counts in `deals_open.csv` at DS3 or later:
- DS3: 61
- DS4: 14
- DS5: 10
- Total DS3+: 61 + 14 + 10 = **85 deals**

Of those 85, deals also present in `gong_calls_by_deal_90d.csv`:
- DS3: 12
- DS4: 6
- DS5: 7
- Total with ≥1 logged call: 12 + 6 + 7 = **25 deals**

Share: 25 / 85 = **0.294 = 29.4%**

So 60 of 85 open DS3+ deals (70.6%) have no row in the provided 90-day Gong table.

Note on owners: `gong_owner_id` in the Gong file is a Gong user id, not a HubSpot owner id, so it was not used to join against `owner_id` in `deals_open.csv`; the mapping check above is on deal id only.
snowflake-sql-trapped-catalogPASS 1.00
Plain-English question: Write one Snowflake query for first-month usage per customer using only the provided, trap-laden catalog.
Prompt — what the model saw
Using ONLY the catalog provided, write one Snowflake SQL query returning, per customer company, for its first calendar month as a customer: unique givers, recognition count, successful redemption count. Read the catalog notes carefully: some tables are stale or unpopulated and must be avoided, and one documented filter (excluding deleted givers) must NOT be applied to historical giving counts. Put a comment above each FROM or JOIN explaining the table choice. Return only fenced SQL.
cost $0.0435 · 22s · in 959 / out 1,776 tok
✓ avoids_stale_deals✓ avoids_unpopulated_live✓ uses_cohort_summary✓ succeeded_filter✓ no_giver_deletion_filter✓ returns_sql

✓ Correct Answer (ground truth)

Plain-English answer: A single correct SQL query that uses the right catalog tables and avoids the trap columns — graded on the exact tables and columns.
The answer in one line: use_table = PRODUCTION.PLG.COMPANY_COHORT_SUMMARY
avoid_stalePRODUCTION.HUBSPOT.DEALS (stale)
avoid_unpopulatedV2_LIVE.OBJECTS_DEALS (unpopulated)
forbidden_filterdeleted-giver exclusion must NOT be applied
outputone fenced SELECT with per-FROM/JOIN table-choice comment
required_filtersucceeded redemptions only
use_tablePRODUCTION.PLG.COMPANY_COHORT_SUMMARY
graded checksavoids_stale_deals · avoids_unpopulated_live · uses_cohort_summary · succeeded_filter · no_giver_deletion_filter · returns_sql

Model's actual answer

show the model's full answer
```sql
-- Per company, for its first calendar month as a customer, this returns
-- M1_USERS as unique givers, M1_GIVES as recognition count, and
-- M1_REDEMPTIONS as successful redemption count.
-- No filter excluding deleted givers is applied, as that filter must NOT
-- be applied to historical giving counts.
SELECT
    COMPANY_ID AS customer_company,
    DATE_TRUNC('month', FIRST_SUB_PAYMENT_DATE) AS first_customer_calendar_month,
    M1_USERS AS unique_givers,
    M1_GIVES AS recognition_count,
    M1_REDEMPTIONS AS successful_redemption_count
-- Table choice: PRODUCTION.PLG.COMPANY_COHORT_SUMMARY is the documented
-- one-row-per-self-serve-company source containing the first-month (M1)
-- user, give, and redemption measures; stale PRODUCTION.HUBSPOT.DEALS and
-- unpopulated V2_LIVE tables are avoided and are not sources for these metrics.
FROM PRODUCTION.PLG.COMPANY_COHORT_SUMMARY
WHERE FIRST_SUB_PAYMENT_DATE IS NOT NULL
ORDER BY customer_company;
```
crm-completeness-enrichmentPASS 1.00
Plain-English question: Audit the CRM for missing and invalid fields, and propose the fixes.
Prompt — what the model saw
Audit the CRM extract for completeness and propose fixes. Deals need owner, stage, amount, close date, why-buys; companies need industry, employee count, HQ country; contacts need email, title, persona. Report completeness percent per field. Identify duplicate company clusters (name variants, shared domain) with a survivor per cluster. List invalid emails and domain mismatches. Fill missing company fields only where the enrichment export has a matching row; where CRM and enrichment disagree list both and recommend a source. Never invent a value. End with the 10 fixes with the most pipeline amount at stake.
cost $0.1437 · 48s · in 7,900 / out 5,195 tok
✓ dup_clusters✓ invalid_emails✓ survivor_language✓ completeness_pct✓ no_invent

✓ Correct Answer (ground truth)

Plain-English answer: 9 companies missing employee count, 6 missing HQ country, 4 invalid emails, 1 domain mismatch, 2 duplicate clusters, and 25 enrichment matches.
The answer in one line: n_contacts = 52; n_companies = 34; enrichment_matches = 25; invalid_email_count = 4; n_dup_clusters = 2; domain_mismatch_count = 1 (+4 more below)
n_contacts52
domain_mismatch_count1
enrichment_matches25
invalid_email_count4
n_companies34
n_dup_clusters2
company_missingindustry=0; employee_count=9; hq_country=6
domain_mismatchesCT-0011
duplicate_clusters['C-0A092931', 'C-0A092932'], ['C-0A092933', 'C-0A092934']
invalid_emailsCT-0010, CT-0080, CT-0081, CT-0192
graded checksdup_clusters · invalid_emails · survivor_language · completeness_pct · no_invent

Model's actual answer

show the model's full answer
Audit — CRM extract completeness and proposed fixes

Scope limitation: No deals file was provided. Only companies.csv (34 rows), contacts.csv (52 rows), and zoominfo_enrichment.csv (25 rows) were provided. Deal completeness and pipeline-at-stake therefore cannot be computed — see last section.

1. Completeness

Deals — NOT COMPUTABLE
  File provided: none.
  owner, stage, amount, close_date, why-buys: no rows, no denominator. Completeness = unknown, not 0%.

Companies (n=34)
  industry:       34/34 = 100.0% present
  employee_count: 25/34 = 73.5% present; 9 missing: C-EC3025, C-96039F, C-44EA29, C-D04904, C-B23205, C-60C75F, C-93C8BF, C-7BBDFA, C-50D386
  hq_country:     28/34 = 82.4% present; 6 missing: C-2D1F1B, C-D73B89, C-44EA29, C-D04904, C-2C60E5, C-EE9FFB
  domain:         34/34 = 100.0% present

  Present-but-nonstandard (counts as present above, needs normalization, not filling):
  - industry variants for the same apparent value: "tech" (C-66D1FC, C-44EA29, C-60C75F, C-0A092932), "Tech " with trailing space (C-425E2A, C-BA969B, C-93C8BF, C-C9BB20), "Technology", "health care" (C-7BBDFA, C-50D386) vs "Healthcare".
  - hq_country variants for the same country: "US", "USA", "United States" all appear. "UK" appears as-is (C-B97B4E, C-332637, C-C9BB20).

Contacts (n=52)
  email:   52/52 = 100.0% non-empty; 48/52 = 92.3% syntactically valid (4 invalid, §3)
  title:   40/52 = 76.9% present; 12 missing: CT-0000, CT-0022, CT-0072, CT-0081, CT-0092, CT-0120, CT-0121, CT-0122, CT-0132, CT-0141, CT-0162, CT-0170
  persona: 39/52 = 75.0% present; 13 missing: CT-0000, CT-0041, CT-0060, CT-0070, CT-0082, CT-0092, CT-0110, CT-0132, CT-0162, CT-0171, CT-0172, CT-0180, CT-0181
  Note: CT-0092 and CT-0132 and CT-0162 are missing both title and persona. All 3 contacts for C-AA8DDA (CT-0120, CT-0121, CT-0122) are missing title.

2. Duplicate company clusters

No company name field was provided, so name-variant duplicates cannot be assessed. Only shared-domain clusters are detectable:

Cluster A — domain acme-corp.com
  C-0A092931: Technology, 500, US
  C-0A092932: tech, 510, USA
  Conflict: employee_count 500 vs 510; industry and country differ only in format.
  Survivor: C-0A092931 (canonical "Technology" spelling). Do not auto-pick 500 vs 510 — unresolved, needs a source decision.

Cluster B — domain globex.io
  C-0A092933: SaaS, 200, US
  C-0A092934: Technology, 200, US
  Conflict: industry SaaS vs Technology (substantive, not formatting).
  Survivor: C-0A092933 (earlier alias). Industry unresolved — neither domain is in the enrichment export, so no third source exists in the data provided.

No other domain repeats. Similar industry labels across different domains (e.g. the many Technology/tech rows) are not duplicate evidence.

3. Invalid emails and domain mismatches

Invalid emails (no domain part — not deliverable as written):
  CT-0010 (C-66D1FC): user0@
  CT-0080 (C-92D97D): user0@
  CT-0081 (C-92D97D): user1@
  CT-0192 (C-425E2A): user2@
  Fix: none proposed — no correct address exists anywhere in the data provided. Re-collect; do not guess the domain in a write, even though the contact's domain column suggests one.

Domain mismatch (email domain ≠ contact/company domain):
  CT-0011 (C-66D1FC): user1@other-domain.com vs domain 66d1fc.com
  This is the only mismatch among syntactically valid emails. All other valid emails match their domain column exactly. Whether other-domain.com is a legitimate alternate domain or an error cannot be determined from the data provided.

4. Enrichment fills and disagreements

Enrichment coverage: 25 rows, matching exactly the first 25 company domains. No enrichment row for: C-BA969B, C-332637, C-93C8BF, C-EE9FFB, C-C9BB20, C-0A092931, C-0A092932, C-0A092933, C-0A092934 (9 companies).

Fills — missing CRM field + matching enrichment row (fill only these):
  employee_count, 8 of 9 missing values fillable (8/9 = 88.9%):
    C-EC3025 ← 400 | C-96039F ← 400 | C-44EA29 ← 400 | C-D04904 ← 400
    C-B23205 ← 400 | C-60C75F ← 400 | C-7BBDFA ← 400 | C-50D386 ← 400
  employee_count, still missing after enrichment: C-93C8BF (93c8bf.com not in enrichment). Leave blank.
  hq_country: 0 of 6 fillable. Every missing-country company either has a blank zi_hq_country (C-2D1F1B, C-D73B89, C-44EA29, C-D04904, C-2C60E5) or no enrichment row (C-EE9FFB). Leave all 6 blank.
  industry / title / persona / email: enrichment contains no such fields beyond industry; no contact fills are possible from it.

Disagreements — both values shown, no value overwritten silently:
  a) employee_count: no disagreements. Every company with both CRM and enrichment values agrees exactly.
  b) hq_country: no substantive disagreements. 9 companies differ only in format — CRM "US"/"USA" vs enrichment "United States": C-66D1FC, C-950043, C-EC3025, C-96039F, C-77A95A, C-B23205, C-E51FB7, C-D0662E, C-425E2A. Recommendation: enrichment as the standard, i.e. normalize all to "United States" — this is a formatting standard, not a fact change.
  c) industry: CRM generic label vs enrichment "Computer Software" (8 companies):
       C-66D1FC: CRM "tech" vs ZI "Computer Software"
       C-EC3025: CRM "Technology" vs ZI "Computer Software"
       C-92D97D: CRM "Technology" vs ZI "Computer Software"
       C-D04904: CRM "Technology" vs ZI "Computer Software"
       C-77A95A: CRM "Technology" vs ZI "Computer Software"
       C-AA8DDA: CRM "Technology" vs ZI "Computer Software"
       C-60C75F: CRM "tech" vs ZI "Computer Software"
       C-425E2A: CRM "Tech " vs ZI "Computer Software"
     Recommendation: enrichment (ZoomInfo) as source for industry where CRM holds only a generic "tech/Technology" label, since "Computer Software" is the more specific classification — but apply it as one explicit mapping decision across all 8, not row-by-row. All other matched industries agree (Manufacturing, Retail, Finance, Healthcare/"health care" capitalization-only).

5. The 10 fixes with the most pipeline amount at stake — CANNOT BE RANKED

No deals file, no deal amounts, and no deal-to-company linkage were provided. Pipeline amount at stake is unknown for every fix below; ranking by it would require inventing numbers. The 10 highest-value fixes identifiable from the data provided, unranked, amount at stake = unknown for each:

  1. Resolve duplicate cluster acme-corp.com (C-0A092931 / C-0A092932), incl. the 500 vs 510 employee conflict.
  2. Resolve duplicate cluster globex.io (C-0A092933 / C-0A092934), incl. SaaS vs Technology.
  3. Re-collect the 4 invalid emails: CT-0010, CT-0080, CT-0081, CT-0192.
  4. Resolve CT-0011's other-domain.com mismatch (C-66D1FC).
  5. Fill employee_count = 400 for the 8 enrichment-matched companies in §4.
  6. Source employee_count for C-93C8BF — no value exists in any file provided.
  7. Source hq_country for the 6 companies in §1 — enrichment cannot fill any of them.
  8. Decide and apply the industry mapping for the 8 "Technology/tech vs Computer Software" conflicts, and normalize "health care"/"Tech " spellings.
  9. Normalize hq_country US/USA/United States to one standard (enrichment spelling recommended).
 10. Backfill contact title (12 missing) and persona (13 missing), starting with CT-0092, CT-0132, CT-0162 (both missing) and the three title-less C-AA8DDA contacts — no source for these exists in the data provided.

To produce the ranked version: provide the deals extract (deal alias, company alias, owner, stage, amount, close date, why-buys) and this section can be re-run with amounts summed per fix.
deal-intelligence 5 tests
closed-lost-classificationPASS 1.00
Plain-English question: Why did we lose the last six months of deals — one primary category per deal, and whose side is the reason on?
Prompt — what the model saw
Classify every deal lost in the last 6 months below into one primary category: pricing, competitor, no decision, timing, product gap, champion left, or other. Also assign a side (Bonusly, buyer, unknown). Then summarize: category counts, the side split, the number of deals where the structured closed_lost_tag clearly disagrees with the free-text reason, and the two patterns most worth acting on. Base everything only on the tag and free-text provided.
cost $0.1652 · 55s · in 5,043 / out 6,796 tok
✓ total_90✓ competitor_top✓ category_counts_present✓ side_split✓ two_patterns

✓ Correct Answer (ground truth)

Plain-English answer: Every closed-lost deal gets exactly one category (pricing, competitor, no decision, timing, product gap, champion left, other) plus a side, per the computed classification.
The answer in one line: total_lost_6mo = 90; top_reason = competitor (26)
total_lost_6mo90
categoriespricing, competitor, no decision, timing, product gap, champion left, other
patternstwo most actionable, tag-vs-text disagreements counted
sideBonusly / buyer / unknown per deal
top_reasoncompetitor (26)
graded checkstotal_90 · competitor_top · category_counts_present · side_split · two_patterns

Model's actual answer

show the model's full answer
Classification rules used, based only on closed_lost_tag + closed_lost_reason_text:
- Primary category = the clearest stated reason in the free text; tag used only when text is vague (e.g. “Timing”, “unresponsive”, “went another direction”).
- Side: Bonusly = text points to Bonusly price/value, product/fit, or a competitor winning on capability/fit. Buyer = text points to buyer budget approval, priorities, timing, internal approval, or leadership change. Unknown = MIA/unresponsive or text too vague to attribute.

Per-deal classification — Deal alias: category / side

Deal-DB0AAC: timing / buyer
Deal-F7F635: competitor / unknown
Deal-AC944F: no decision / unknown
Deal-214060: no decision / unknown
Deal-91A056: timing / buyer
Deal-29326C: timing / buyer
Deal-5DB9B0: other / unknown
Deal-831B7B: timing / buyer
Deal-F97C37: competitor / Bonusly
Deal-13E9CF: no decision / buyer
Deal-39E25C: timing / buyer
Deal-7ED004: pricing / buyer
Deal-21B045: no decision / unknown
Deal-B3ABED: timing / buyer
Deal-422BA6: competitor / Bonusly
Deal-ED9AE7: timing / buyer
Deal-988493: no decision / unknown
Deal-381C8C: competitor / unknown
Deal-F308CA: no decision / unknown
Deal-F1E8A6: competitor / unknown
Deal-B6AC09: timing / buyer
Deal-70F704: no decision / unknown
Deal-E6E80A: timing / buyer
Deal-B038F0: timing / buyer
Deal-4664E1: no decision / unknown
Deal-175756: timing / buyer
Deal-E74A73: no decision / buyer
Deal-DDAB52: competitor / Bonusly
Deal-ACE061: competitor / unknown
Deal-BB78F3: timing / buyer
Deal-D48E0B: no decision / unknown
Deal-15DA99: timing / buyer
Deal-F4AF5D: timing / buyer
Deal-79B7A1: timing / buyer
Deal-583ADB: no decision / unknown
Deal-8E27DA: product gap / Bonusly
Deal-2D2F8D: competitor / unknown
Deal-E0441F: no decision / unknown
Deal-7CB44D: no decision / unknown
Deal-0F96AA: competitor / unknown
Deal-1BCA50: pricing / Bonusly
Deal-7CC678: competitor / unknown
Deal-FAC17C: no decision / buyer
Deal-242273: competitor / Bonusly
Deal-50E5D8: no decision / buyer
Deal-A2C349: competitor / Bonusly
Deal-9F176A: timing / buyer
Deal-7B2236: pricing / Bonusly
Deal-AFA56C: no decision / unknown
Deal-C7156E: competitor / unknown
Deal-C33D91: pricing / buyer
Deal-9048EB: no decision / unknown
Deal-5E64CE: timing / buyer
Deal-8A0992: competitor / Bonusly
Deal-D0C698: competitor / unknown
Deal-69CF3D: timing / buyer
Deal-ECBF89: timing / buyer
Deal-3618CC: product gap / Bonusly
Deal-EECC02: competitor / unknown
Deal-5AD03E: product gap / Bonusly
Deal-D1A623: timing / buyer
Deal-413C56: no decision / buyer
Deal-47F1A1: competitor / unknown
Deal-BF2A98: competitor / unknown
Deal-2A292B: no decision / buyer
Deal-D1AABF: no decision / unknown
Deal-FEDBCB: timing / buyer
Deal-1E7DA9: competitor / unknown
Deal-2BBA21: no decision / unknown
Deal-286F9C: competitor / unknown
Deal-7FBAC6: no decision / buyer
Deal-369281: competitor / unknown
Deal-386F6E: no decision / unknown
Deal-9FCD0D: competitor / Bonusly
Deal-55867E: no decision / buyer
Deal-DAFB82: pricing / buyer
Deal-2FEDDB: timing / buyer
Deal-64B19A: competitor / unknown
Deal-3F86A0: no decision / unknown
Deal-096750: no decision / unknown
Deal-F325A5: champion left / buyer
Deal-ABD14C: no decision / buyer
Deal-79E61A: no decision / unknown
Deal-8A119B: pricing / buyer
Deal-AE7C4E: no decision / unknown
Deal-DAB4F1: no decision / unknown
Deal-B4B50F: no decision / unknown
Deal-981AD4: product gap / Bonusly
Deal-DC77FE: competitor / Bonusly
Deal-5885B9: no decision / unknown

Summary

Total deals classified: 90

Category counts:
- no decision: 32 / 90 = 35.6%
- competitor: 24 / 90 = 26.7%
- timing: 22 / 90 = 24.4%
- pricing: 6 / 90 = 6.7%
- product gap: 4 / 90 = 4.4%
- champion left: 1 / 90 = 1.1%
- other: 1 / 90 = 1.1%
Check: 32 + 24 + 22 + 6 + 4 + 1 + 1 = 90

Side split:
- unknown: 40 / 90 = 44.4%
- buyer: 36 / 90 = 40.0%
- Bonusly: 14 / 90 = 15.6%
Check: 40 + 36 + 14 = 90

Structured tag clearly disagrees with free text: 7 deals

Counted only where the text clearly contradicts, replaces, or negates the tag’s primary meaning — not where text is merely vague:
1. Deal-13E9CF — tag “Doing nothing/Not a priority/Cost”; text: “Not a budget issue - R&R program has been deprioritized”
2. Deal-1BCA50 — tag “Competitor”; text: “It was mostly about the budget…” with another vendor only secondary
3. Deal-70F704 — tag “Lost DM”; text: only anniversary-award scope + “have been MIA”, no decision-maker loss stated
4. Deal-5E64CE — tag “Doing nothing/Not a priority/Cost”; text: Nectar agreement “through October 2027”, plan to reach out near contract end — timing/contract, not doing nothing
5. Deal-3618CC — tag “Lost DM”; text: “Wanted Surveys” — product gap
6. Deal-5AD03E — tag “Competitor”; text: “Wanted more defined budget access” — product gap, no competitor stated
7. Deal-55867E — tag “Lost- Timing (1 year or more)”; text: “I don't think we'll be moving forward with Bonusly at this time” — no 1-year+ timing stated

7 / 90 = 7.8% clear tag-vs-text disagreements. Missing data note: many more texts are too vague to confirm or contradict the tag — that is counted under side = unknown, not as disagreement.

Two patterns most worth acting on

1. No-decision / MIA is the largest bucket and is mostly unattributable.
32 no-decision deals, of which 23 are side = unknown, driven by “unresponsive”, “MIA”, “No contact… ignored outreach”. Combined with 40/90 unknown-side overall, the data cannot tell whether these were winnable. Act on qualification and exit discipline earlier: no closed-lost without a stated buyer reason, last-contact date, and a named blocker.

2. Timing is a recycle pool, not a dead pool.
22 timing deals, all buyer-side, with explicit return points in the text — e.g. Deal-91A056 “reconnect early in 2027”, Deal-E6E80A “pushed into early 2027”, Deal-B3ABED “revisit… likely in Q2 next year”, Deal-5E64CE contract “through October 2027”. These need dated re-open tasks tied to the stated window, separate from true no-decision losses.
pipeline-tieringPASS 1.00
Plain-English question: Score all 156 open deals into exactly one of LOCK / ACTION / BUILD / REVIVE / WATCH / RISKY.
Prompt — what the model saw
Score the open pipeline below on these signals: forecast category, stage, meetings_30d, contact count, and engagement recency. Assign exactly one tier per deal: LOCK, ACTION, BUILD, REVIVE, WATCH, or RISKY (RISKY = the forecast category disagrees with the engagement evidence). Never tier a deal LOCK if it has zero meetings_30d. Do NOT list every deal. Return ONLY a compact JSON object with exactly these keys: - "tier_counts": object mapping each tier to the number of deals in it (must sum to the total deal count), - "tier_examples": object mapping each tier to up to 3 example deal aliases from the data, - "risky_deals": array of the deal aliases you tiered RISKY, - "lock_violations": number of deals tiered LOCK that have zero meetings_30d (should be 0), - "pipeline_shape": one short paragraph on the overall pipeline shape. Note: inbound_emails_30d is 0 for every row due to a data defect, so treat meetings_30d as the inbound signal.
cost $0.4444 · 47s · in 81,673 / out 3,615 tok
✓ json_returned✓ tier_vocabulary✓ tier_counts_sum✓ risky_deals_listed✓ no_lock_violation_claimed

✓ Correct Answer (ground truth)

Plain-English answer: All 156 open deals tiered exactly once across the six tiers using forecast category, stage, meetings, contacts, and recency.
The answer in one line: total = 156
total156
graded checksjson_returned · tier_vocabulary · tier_counts_sum · risky_deals_listed · no_lock_violation_claimed

Model's actual answer

show the model's full answer
{"tier_counts": {"LOCK": 6, "ACTION": 20, "BUILD": 54, "REVIVE": 5, "WATCH": 36, "RISKY": 35}, "tier_examples": {"LOCK": ["Deal-25F752", "Deal-D348E1", "Deal-C26D20"], "ACTION": ["Deal-A5E80A", "Deal-1FC049", "Deal-499BF6"], "BUILD": ["Deal-6787C2", "Deal-D73B89", "Deal-012CB1"], "REVIVE": ["Deal-2D1F1B", "Deal-F0EBBB", "Deal-3EED2C"], "WATCH": ["Deal-66D1FC", "Deal-950043", "Deal-B23205"], "RISKY": ["Deal-E53952", "Deal-5408B0", "Deal-9AAE5F"]}, "risky_deals": ["Deal-E53952", "Deal-5408B0", "Deal-9AAE5F", "Deal-547B2B", "Deal-B7EBD1", "Deal-A2B47C", "Deal-2465CE", "Deal-C61CF7", "Deal-62D607", "Deal-584EE5", "Deal-C6D97A", "Deal-7B3B0F", "Deal-F9A08A", "Deal-0660B4", "Deal-FD9F4E", "Deal-BA571A", "Deal-FC22A3", "Deal-7BBDFA", "Deal-60C2C2", "Deal-4A13AD", "Deal-8AD4A5", "Deal-15D24F", "Deal-9D0060", "Deal-690476", "Deal-635B8E", "Deal-ED725A", "Deal-55164C", "Deal-3BA5EA", "Deal-5FDCE4", "Deal-F336B6", "Deal-5EED42", "Deal-BA3DDC", "Deal-7599B8", "Deal-F9A3C1", "Deal-FA32A0"], "lock_violations": 0, "pipeline_shape": "156 deals total (6+20+54+5+36+35=156), heavily early-stage and PIPELINE-weighted (105/156 PIPELINE, 132/156 at DS1-DS3) with thin meeting evidence overall (101/156 with zero meetings_30d, only 55/156 with >=1); the late/high-forecast book is split sharply, with only 6 LOCK deals versus 35 RISKY deals where COMMIT/BEST_CASE has zero meetings_30d, so most value sits in BUILD/WATCH (90/156) rather than in substantiated late-stage pipeline."}
call-transcript-extractionPASS 1.00
Plain-English question: Extract the CRM write-back fields from each call transcript as JSON.
Prompt — what the model saw
For each transcript, extract CRM write-back fields as JSON: why-buys (prospect statements only), pain points, stakeholders from the speaker list, budget signal (prospect-stated or null), timeline signal, competitor mentioned (only if the prospect raised it), next step (only if explicitly agreed), objections, confidence. Never fill a field from what the rep said or hoped.
cost $0.0501 · 17s · in 1,791 / out 1,940 tok
✓ json_emitted✓ budget_signal✓ null_discipline✓ stakeholders✓ competitor_prospect_only

✓ Correct Answer (ground truth)

Plain-English answer: Per transcript: prospect-only why-buys, pain points, stakeholders, budget and timeline signals, and competitors — e.g. TX-001: '$40k earmarked', live before open enrollment.
The answer in one line: rep = Alex Franklin
allowed_newTX-001, TX-002, TX-003, TX-004, TX-005, TX-006, Achievers, Bucketlist, Nectar
per_transcript{"TX-001": {"why_buys": ["automating anniversary and birthday awards"], "pain_points": ["HR team of three cannot keep up with it manually", "track everything in a spreadsheet, and people slip through
repAlex Franklin
rep_sourced_nulls{"TX-003": ["budget_signal \u2014 only the rep mentioned pricing/budget"], "TX-004": ["next_step \u2014 rep proposed, prospect said 'no promises'"], "TX-006": ["budget_signal \u2014 rep decoy, prospec
graded checksjson_emitted · budget_signal · null_discipline · stakeholders · competitor_prospect_only

Model's actual answer

show the model's full answer
[
  {
    "transcript_id": "TX-001",
    "deal_alias": "Deal-CFE7F4",
    "why_buys": [
      "Automating anniversary and birthday awards"
    ],
    "pain_points": [
      "HR team of three cannot keep up with anniversary and birthday awards manually",
      "Everything is tracked in a spreadsheet, and people slip through the cracks"
    ],
    "stakeholders": [
      "Prospect (VP People)",
      "Prospect (HR Admin)"
    ],
    "budget_signal": "About $40k earmarked for engagement tools this fiscal year",
    "timeline_signal": "Ideally live before open enrollment in November",
    "competitor_mentioned": "Achievers",
    "next_step": "Security review on September 12",
    "objections": [
      "Need SSO and audit logs for IT to sign off"
    ],
    "confidence": "high"
  },
  {
    "transcript_id": "TX-002",
    "deal_alias": "Deal-70BB30",
    "why_buys": [
      "Tie recognition to retention for hourly workforce"
    ],
    "pain_points": [
      "Regretted turnover for hourly workforce is over 30%"
    ],
    "stakeholders": [
      "Prospect (Head of Total Rewards)",
      "Prospect (CFO)"
    ],
    "budget_signal": "$25k pilot budget approved by Finance for this quarter",
    "timeline_signal": "Decision by end of September",
    "competitor_mentioned": null,
    "next_step": "Send pilot agreement; prospect will route it to legal this week",
    "objections": [
      "Integration with Workday has to be rock solid — CFO condition"
    ],
    "confidence": "high"
  },
  {
    "transcript_id": "TX-003",
    "deal_alias": "Deal-530B50",
    "why_buys": [
      "Make recognition visible across 12 retail locations"
    ],
    "pain_points": [
      "Store managers have zero budget autonomy for on-the-spot recognition today"
    ],
    "stakeholders": [
      "Prospect (People Ops Manager)"
    ],
    "budget_signal": null,
    "timeline_signal": "No rush until Q1",
    "competitor_mentioned": "Bucketlist",
    "next_step": "Schedule a call with the CEO; prospect will send two times",
    "objections": [
      "CEO has to be sold first — she decides anything people-related"
    ],
    "confidence": "high"
  },
  {
    "transcript_id": "TX-004",
    "deal_alias": "Deal-180D02",
    "why_buys": [
      "Consolidate three separate recognition tools into one"
    ],
    "pain_points": [
      "Paying for three tools and none of them talk to the HRIS"
    ],
    "stakeholders": [
      "Prospect (VP People)",
      "Prospect (IT Security Lead)"
    ],
    "budget_signal": "Under $15k annually can be approved by VP People without going to the board",
    "timeline_signal": "Procurement cycle runs six to eight weeks minimum",
    "competitor_mentioned": null,
    "next_step": null,
    "objections": [
      "Security review took three months for last vendor — hesitation",
      "Procurement cycle runs six to eight weeks minimum"
    ],
    "confidence": "high"
  },
  {
    "transcript_id": "TX-005",
    "deal_alias": "Deal-F8767A",
    "why_buys": [
      "Automate service milestones",
      "Analytics on recognition equity across departments"
    ],
    "pain_points": [
      "Night-shift teams feel invisible — engagement scores run 20 points lower"
    ],
    "stakeholders": [
      "Prospect (HR Director)",
      "Prospect (People Ops Coordinator)"
    ],
    "budget_signal": "$12k approved under engagement line",
    "timeline_signal": "Need running before January all-hands",
    "competitor_mentioned": "Nectar",
    "next_step": "Present to exec team on October 2",
    "objections": [
      "Exec team is skeptical after a failed rollout two years ago",
      "Mid-pilot with Nectar — vendor must beat that experience"
    ],
    "confidence": "high"
  },
  {
    "transcript_id": "TX-006",
    "deal_alias": "Deal-EE195F",
    "why_buys": [
      "Cut admin time on service awards"
    ],
    "pain_points": [
      "HR Manager personally spends five hours a month ordering and shipping plaques"
    ],
    "stakeholders": [
      "Prospect (HR Manager)"
    ],
    "budget_signal": null,
    "timeline_signal": "Fiscal year resets in February, so a Q1 start is realistic",
    "competitor_mentioned": null,
    "next_step": "Send one-page overview; prospect will forward it to COO this week",
    "objections": [
      "COO usually prefers building things in-house",
      "Comparison is against doing it internally"
    ],
    "confidence": "high"
  }
]
next-to-closePASS 1.00
Plain-English question: Which 3 deals are closest to signature right now?
Prompt — what the model saw
Which 3 deals are closest to signature right now? Use the CRM data and the Slack thread; Slack often has the real status before the CRM does. For each: deal, amount, why it is close, what is left. Under 250 words.
cost $0.0306 · 16s · in 1,106 / out 1,105 tok
✓ exactly_three✓ top3_named✓ slack_first✓ blocked_excluded

✓ Correct Answer (ground truth)

Plain-English answer: Deal-547B2B ($11,200), Deal-B7EBD1 ($9,000), and Deal-403845 ($9,000) — all DS5 COMMIT deals closing September 10-11, with what is left on each.
blocked_excludeddeal_alias=Deal-2465CE; reason=champion left + procurement freeze (Slack #sales-sparring 2026-09-04)
exactly_three1
ranking_ruleSlack signature-imminent deal ranks first; remaining ranked by stage (DS5>DS4), forecast (COMMIT>BEST_CASE>PIPELINE), close_date asc, amount desc
top3rank=1; deal_alias=Deal-547B2B; amount=11200; stage=DS5; forecast_category=COMMIT; close_date=2026-09-11 | rank=2; deal_alias=Deal-B7EBD1; amount=9000; stage=DS5; forecast_category=COMMIT; close_date=2026-09-10 | rank=3; deal_alias=Deal-403845; amount=9000; stage=DS5; forecast_category=COMMIT; close_date=2026-09-11
graded checksexactly_three · top3_named · slack_first · blocked_excluded

Model's actual answer

show the model's full answer
Closest to signature:

1. **Deal-547B2B — $11,200**
   - Why: CRM: DS5, COMMIT, close 2026-09-11. Slack (2026-09-04, Alex Franklin): “redlines came back clean,” “signing page is out,” VP People “signing tomorrow.”
   - Left: The signature itself.

2. **Deal-403845 — $9,000**
   - Why: CRM: DS5, COMMIT, close 2026-09-11. Slack (2026-09-04, Dana Mercer): “moving fine,” “order form is with their finance team.”
   - Left: Finance team’s completion of the order form. No signature date stated.

3. **Deal-B7EBD1 — $9,000**
   - Why: CRM: DS5, COMMIT, earliest close date in the file, 2026-09-10. No Slack blocker reported for this deal.
   - Left: Not stated in the provided data — no Slack update for Deal-B7EBD1.

Not included despite CRM DS5/COMMIT/2026-09-10: Deal-2465CE ($5,400) — Slack says champion left, procurement froze new vendors, owner is “pulling it out of commit,” “blocked,” realistically Q4.
product-gap-detectionPASS 1.00
Plain-English question: Which prospect complaints are true product gaps versus rollout or enablement issues?
Prompt — what the model saw
Surface true product gaps raised by prospects in active deals. For each candidate: quote the prospect line, classify as real gap, plan gate, or rollout/enablement issue using the product docs excerpt, and name the deal and amount. Only real gaps go in the summary. Rep language does not count as prospect voice.
cost $0.0278 · 14s · in 719 / out 1,042 tok
✓ real_gaps✓ plan_gate_classified✓ enablement_classified✓ gap_summary

✓ Correct Answer (ground truth)

Plain-English answer: Two real gaps (SCIM provisioning, ADP Workforce Now integration) separated from plan gates and enablement issues, each with the prospect quote and deal.
allowed_newTG-001, TG-002, TG-003, TG-004, TG-005, SCIM user provisioning, ADP Workforce Now integration, Custom report builder
enablement_issuestranscript_id=TG-004; deal_alias=Deal-84DBA6; amount=16000; feature=Slack integration recognition adoption; quote=We turned on the Slack integration in July, but half our managers still haven't given a single recognition — nobody ever trained them on it.; why_enablement=Slack integration exists on all plans; managers were never trained
excluded_rep_voicetranscript_id=TG-005; line=Full transparency — we don't have a native mobile app for hourly workers yet, though it's on the roadmap.; reason=rep voice, not prospect-raised
plan_gatestranscript_id=TG-003; deal_alias=Deal-CFE7F4; amount=32000; feature=Custom report builder; tier_required=Enterprise; quote=I tried to build a custom report combining recognition frequency with turnover and couldn't — is that possible on our current plan?
real_gapstranscript_id=TG-001; deal_alias=Deal-EC3025; amount=62000; feature=SCIM user provisioning; quote=We need SCIM user provisioning — our security team will not approve any tool that requires manual account management.; why_real=not listed in product docs at any tier | transcript_id=TG-002; deal_alias=Deal-D0D6B5; amount=23800; feature=ADP Workforce Now integration; quote=Our payroll runs on ADP Workforce Now and I don't see ADP anywhere in your integration list — that's a dealbreaker for us.; why_real=supported HRIS list is Workday/BambooHR/Gusto/Rippling only
graded checksreal_gaps · plan_gate_classified · enablement_classified · gap_summary

Model's actual answer

show the model's full answer
Candidate review — prospect voice only

1. Deal-EC3025
Prospect line: "We need SCIM user provisioning — our security team will not approve any tool that requires manual account management."
Classification: Real gap
Basis: product_docs.md states "SCIM user provisioning and ADP Workforce Now integrations are NOT currently listed as supported capabilities." It also does not appear in the Plans and tiers table.
Amount: Not provided in the data.

2. Deal-D0D6B5
Prospect line: "Our payroll runs on ADP Workforce Now and I don't see ADP anywhere in your integration list — that's a dealbreaker for us."
Classification: Real gap
Basis: product_docs.md states "SCIM user provisioning and ADP Workforce Now integrations are NOT currently listed as supported capabilities." Supported HRIS integrations are listed only as "Workday, BambooHR, Gusto, Rippling (Pro and above)."
Amount: Not provided in the data.

3. Deal-CFE7F4
Prospect line: "I tried to build a custom report combining recognition frequency with turnover and couldn't — is that possible on our current plan?"
Classification: Plan gate
Basis: product_docs.md lists "Custom report builder" as Enterprise-only in the Plans and tiers table.
Amount: Not provided in the data.

4. Deal-84DBA6
Prospect line: "We turned on the Slack integration in July, but half our managers still haven't given a single recognition — nobody ever trained them on it."
Classification: Rollout/enablement issue
Basis: product_docs.md lists "Slack and Microsoft Teams integration" as "yes" on Core, Pro, and Enterprise. The prospect line attributes the problem to lack of training, not a missing capability.
Amount: Not provided in the data.

Excluded — not prospect voice

Deal-36C33F: The mobile-app line is spoken by Alex Franklin: "Full transparency — we don't have a native mobile app for hourly workers yet, though it's on the roadmap." Rep language does not count as prospect voice. The prospect line is: "Good to know. The web version should be fine for our office staff for now."
Amount: Not provided in the data.

Arithmetic

Prospect-raised candidates reviewed: 4 (TG-001, TG-002, TG-003, TG-004)
Real gaps: 2 of 4
Plan gates: 1 of 4
Rollout/enablement issues: 1 of 4
Excluded for speaker: 1 of 5 total transcript records (TG-005)

Summary — real gaps only

- Deal-EC3025 — SCIM user provisioning. Prospect: "We need SCIM user provisioning — our security team will not approve any tool that requires manual account management." Amount: not provided.
- Deal-D0D6B5 — ADP Workforce Now integration. Prospect: "Our payroll runs on ADP Workforce Now and I don't see ADP anywhere in your integration list — that's a dealbreaker for us." Amount: not provided.

No deal amounts were provided in transcripts_gaps.csv or product_docs.md, so no amount totals can be calculated.
rep-performance 5 tests
stale-pipeline-by-rep0.83
Plain-English question: Which open deals have had no email, call, or meeting in the last 7 days?
Prompt — what the model saw
The snapshot date is 2026-09-05. Using the open deals and the engagements-by-deal table, list every open deal with no logged email, call, or meeting in the last 7 days. Compute recency from the engagements table's last_email/last_call/last_meeting fields, NOT from any date field on the deal (the deal's last_contacted_field is unreliable). For each stale deal give: deal alias, owner name, stage, amount, days since last contact. Group by owner, order by amount descending within each owner. Then state, per owner, how many stale deals they carry and the total stale amount. (Do not write outreach text.)
cost $0.4084 · 37s · in 77,845 / out 4,052 tok
✓ stale_recall_ge_80✓ grouped_by_owner✓ bryce_top_amount✓ cole_stale_amt✓ dana_stale_amt✗ stale_count

✓ Correct Answer (ground truth)

Plain-English answer: 65 open deals are stale; the largest is Deal-2D1F1B at $240,000, with Bryce Harmon carrying the most stale dollars at $626,243.
The answer in one line: jake_amt = 626,243; jenn_amt = 261,645; kevin_amt = 252,905.03; top_amount = 240,000; stale_count = 65; top_alias = Deal-2D1F1B
jake_amt626,243
jenn_amt261,645
kevin_amt252,905.03
stale_count65
top_amount240,000
top_aliasDeal-2D1F1B
graded checksstale_recall_ge_80 · grouped_by_owner · bryce_top_amount · cole_stale_amt · dana_stale_amt · stale_count

Model's actual answer

show the model's full answer
Snapshot: 2026-09-05

Method: For each open deal, last contact = MAX(last_email, last_call, last_meeting) from engagements_by_deal_90d.csv. Days since last contact = 2026-09-05 − last contact. Stale = days since last contact > 7, i.e. last contact on or before 2026-08-28. The deal field last_contacted_field was not used.

Bryce Harmon
- Deal-2D1F1B | DS1 | 240000 | 81 days (last contact 2026-06-16)
- Deal-66D1FC | DS1 | 99000 | 16 days (2026-08-20)
- Deal-950043 | DS1 | 70000 | 19 days (2026-08-17)
- Deal-B23205 | DS1 | 45000 | 16 days (2026-08-20)
- Deal-7BBDFA | DS3 | 37440 | 46 days (2026-07-21)
- Deal-332637 | DS2 | 36000 | 9 days (2026-08-27)
- Deal-1BEEBF | DS1 | 31500 | 19 days (2026-08-17)
- Deal-C5658B | DS1 | 23400 | 16 days (2026-08-20)
- Deal-40522D | DS3 | 21000 | 19 days (2026-08-17)
- Deal-F0EBBB | DS3 | 11400 | 24 days (2026-08-12)
- Deal-E25A09 | DS1 | 6000 | 9 days (2026-08-27)
- Deal-C9C286 | DS2 | 5502 | 9 days (2026-08-27)
- Deal-012CB1 | DS1 | 1 | 23 days (2026-08-13)

Dana Mercer
- Deal-44EA29 | DS2 | 60000 | 10 days (2026-08-26)
- Deal-E51FB7 | DS2 | 43875 | 12 days (2026-08-24)
- Deal-B42F46 | DS1 | 27000 | 19 days (2026-08-17)
- Deal-BA3DDC | DS3 | 23400 | 15 days (2026-08-21)
- Deal-9DDE86 | DS2 | 20000 | 15 days (2026-08-21)
- Deal-215CCA | DS3 | 18900 | 17 days (2026-08-19)
- Deal-5EED42 | DS3 | 16250 | 11 days (2026-08-25)
- Deal-57887A | DS2 | 15000 | 8 days (2026-08-28)
- Deal-B7EBD1 | DS5 | 9000 | 16 days (2026-08-20)
- Deal-3974EB | DS4 | 9000 | 8 days (2026-08-28)
- Deal-F40F04 | DS2 | 8100 | 15 days (2026-08-21)
- Deal-87DDD1 | DS1 | 5000 | 19 days (2026-08-17)
- Deal-F336B6 | DS3 | 4200 | 15 days (2026-08-21)
- Deal-0660B4 | DS4 | 1920 | 16 days (2026-08-20)

Alex Franklin
- Deal-CC08D1 | DS1 | 24000 | 16 days (2026-08-20)
- Deal-E73427 | DS3 | 18000 | 10 days (2026-08-26)
- Deal-885F45 | DS2 | 9300 | 12 days (2026-08-24)
- Deal-C2FF3C | DS1 | 8316 | 10 days (2026-08-26)
- Deal-3EED2C | DS2 | 7200 | days since last contact: data missing — no row in engagements_by_deal_90d.csv
- Deal-0D2F7A | DS3 | 5100 | 12 days (2026-08-24)
- Deal-6C60D4 | DS3 | 4800 | 12 days (2026-08-24)
- Deal-13FEBD | DS2 | 4680 | 12 days (2026-08-24)
- Deal-9D0060 | DS3 | 3840 | 12 days (2026-08-24)
- Deal-690476 | DS2 | 3600 | 18 days (2026-08-18)
- Deal-C6D97A | DS4 | 3240 | 8 days (2026-08-28)
- Deal-EE195F | DS3 | 3120 | 8 days (2026-08-28)
- Deal-278DEC | DS3 | 2700 | 8 days (2026-08-28)
- Deal-635B8E | DS3 | 2600 | 18 days (2026-08-18)
- Deal-6883F3 | DS1 | 2400 | 16 days (2026-08-20)
- Deal-4A13AD | DS3 | 2160 | 26 days (2026-08-10)
- Deal-F67D31 | DS2 | 1800 | 8 days (2026-08-28)
- Deal-5FDCE4 | DS3 | 1600 | 12 days (2026-08-24)
- Deal-BA571A | DS4 | 1080 | 18 days (2026-08-18)

Cole Ingram
- Deal-D04904 | DS2 | 58529.25 | 11 days (2026-08-25)
- Deal-B25F40 | DS3 | 40000 | 8 days (2026-08-28)
- Deal-813836 | DS2 | 32175 | 11 days (2026-08-25)
- Deal-1BA595 | DS2 | 31750 | 11 days (2026-08-25)
- Deal-CFE1E8 | DS3 | 18000 | 11 days (2026-08-25)
- Deal-CD47A6 | DS2 | 12168 | 11 days (2026-08-25)
- Deal-627646 | DS3 | 11193 | 11 days (2026-08-25)
- Deal-FF809F | DS2 | 7781.2 | 11 days (2026-08-25)
- Deal-AF932D | DS2 | 7225.4 | 11 days (2026-08-25)
- Deal-A71728 | DS2 | 6947.5 | 11 days (2026-08-25)
- Deal-8BC9F5 | DS2 | 5616 | 10 days (2026-08-26)
- Deal-175395 | DS3 | 4779.88 | 11 days (2026-08-25)
- Deal-481E24 | DS3 | 4140 | 10 days (2026-08-26)
- Deal-C7F9BF | DS2 | 3360 | 11 days (2026-08-25)
- Deal-2F3A66 | DS3 | 3334.8 | 11 days (2026-08-25)
- Deal-342E96 | DS2 | 2700 | 24 days (2026-08-12)
- Deal-E568D5 | DS3 | 1875 | 11 days (2026-08-25)
- Deal-FD9F4E | DS5 | 1330 | 10 days (2026-08-26)

Elena Sinclair
- Deal-57FF13 | DS1 | 2100 | days since last contact: data missing — no row in engagements_by_deal_90d.csv

Per-owner stale totals
- Bryce Harmon: 13 stale deals; 240000+99000+70000+45000+37440+36000+31500+23400+21000+11400+6000+5502+1 = 626243
- Dana Mercer: 14 stale deals; 60000+43875+27000+23400+20000+18900+16250+15000+9000+9000+8100+5000+4200+1920 = 261645
- Alex Franklin: 19 stale deals; 24000+18000+9300+8316+7200+5100+4800+4680+3840+3600+3240+3120+2700+2600+2400+2160+1800+1600+1080 = 109536
- Cole Ingram: 18 stale deals; 58529.25+40000+32175+31750+18000+12168+11193+7781.2+7225.4+6947.5+5616+4779.88+4140+3360+3334.8+2700+1875+1330 = 252905.03
- Elena Sinclair: 1 stale deal; 2100 = 2100
- Farid Osman: 0 stale deals; 0
activity-mix-vs-outcomePASS 1.00
Plain-English question: Per rep, what did the last 30 days of emails, calls, and meetings look like next to the deals that advanced?
Prompt — what the model saw
The snapshot date is 2026-09-05. For each sales rep (owner), compute the last-30-day totals of emails, calls, and meetings from the engagements table, and the number of deals that entered DS2 in the last 30 days (t_ds2 within 30 days of the snapshot). Report per rep: the activity mix as percentages (emails/calls/meetings share of that rep's total activities), total activities per DS2 entry, and rank the reps by that efficiency ratio (lowest activities per DS2 entry = most efficient). Name the most efficient rep and the highest-volume rep, and note if they differ. Do not attribute cause.
cost $0.4187 · 31s · in 83,375 / out 3,472 tok
✓ alex_ds2_30d✓ alex_volume_high✓ efficient_named✓ mix_present✓ no_causal_claim

✓ Correct Answer (ground truth)

Plain-English answer: Alex Franklin is the standout signal with 18 deals entering DS2 in 30 days (next closest: Bryce Harmon with 4), alongside per-rep activity totals.
The answer in one line: alex_ds2_30d = 18; bryce_ds2_30d = 4; note = legacy alias justin_ds2=4 actually held Bryce Harmon count; Alex Franklin entered 18 deals into DS2 in the 30d window
alex_ds2_30d18
bryce_ds2_30d4
notelegacy alias justin_ds2=4 actually held Bryce Harmon count; Alex Franklin entered 18 deals into DS2 in the 30d window
graded checksalex_ds2_30d · alex_volume_high · efficient_named · mix_present · no_causal_claim

Model's actual answer

show the model's full answer
Snapshot: 2026-09-05  
Last-30-day DS2 window used: 2026-08-06 to 2026-09-05 inclusive, i.e. `2026-09-05 - 30 days = 2026-08-06 ≤ t_ds2 ≤ 2026-09-05`.  
Activities are the sums of `emails_30d`, `calls_30d`, and `meetings_30d` in engagements_by_deal_90d.csv, joined to deals_open.csv by `deal_id` and grouped by `owner_id`.

Per-rep totals, mix, and efficiency

Owner names are from owners.csv as provided.

1. Alex Franklin (84342457) — Rank 1, most efficient
- Emails = 307, Calls = 36, Meetings = 41
- Total activities = 307 + 36 + 41 = 384
- Activity mix: emails 307/384 = 79.95%, calls 36/384 = 9.38%, meetings 41/384 = 10.68%
- DS2 entries in window = 18: Deal-EE195F (2026-08-06), Deal-D9A72E (2026-08-06), Deal-7FA0C3 (2026-08-07), Deal-E531A6 (2026-08-07), Deal-36C33F (2026-08-11), Deal-D1E6C2 (2026-08-11), Deal-317E6F (2026-08-12), Deal-4F775F (2026-08-17), Deal-F436DA (2026-08-19), Deal-CA5E44 (2026-08-24), Deal-46988D (2026-08-26), Deal-5296C9 (2026-08-28), Deal-898FC5 (2026-08-28), Deal-E73427 (2026-08-28), Deal-403845 (2026-09-02), Deal-92D97D (2026-09-02), Deal-1FC049 (2026-09-03), Deal-3EED2C (2026-09-03)
- Total activities per DS2 entry = 384 / 18 = 21.33

2. Bryce Harmon (119337721) — Rank 2
- Emails = 162, Calls = 0, Meetings = 43
- Total activities = 162 + 0 + 43 = 205
- Activity mix: emails 162/205 = 79.02%, calls 0/205 = 0.00%, meetings 43/205 = 20.98%
- DS2 entries in window = 4: Deal-1CCE5C (2026-08-06), Deal-25F752 (2026-08-10), Deal-CA7DC0 (2026-08-12), Deal-D73B89 (2026-09-03)
- Total activities per DS2 entry = 205 / 4 = 51.25

3. Cole Ingram (83155924) — Rank 3
- Emails = 96, Calls = 14, Meetings = 1
- Total activities = 96 + 14 + 1 = 111
- Activity mix: emails 96/111 = 86.49%, calls 14/111 = 12.61%, meetings 1/111 = 0.90%
- DS2 entries in window = 2: Deal-1BA595 (2026-08-12), Deal-42326B (2026-08-26)
- Total activities per DS2 entry = 111 / 2 = 55.50

4. Farid Osman (716654662) — Rank 4
- Emails = 38, Calls = 0, Meetings = 34
- Total activities = 38 + 0 + 34 = 72
- Activity mix: emails 38/72 = 52.78%, calls 0/72 = 0.00%, meetings 34/72 = 47.22%
- DS2 entries in window = 1: Deal-499BF6 (2026-08-26)
- Total activities per DS2 entry = 72 / 1 = 72.00

5. Dana Mercer (83155923) — Rank 5
- Emails = 84, Calls = 18, Meetings = 11
- Total activities = 84 + 18 + 11 = 113
- Activity mix: emails 84/113 = 74.34%, calls 18/113 = 15.93%, meetings 11/113 = 9.73%
- DS2 entries in window = 1: Deal-57887A (2026-08-07)
- Total activities per DS2 entry = 113 / 1 = 113.00

Unranked — Elena Sinclair (701163055)
- Emails = 0, Calls = 0, Meetings = 0; Total = 0 + 0 + 0 = 0
- Activity mix: undefined, because total activities = 0 (0/0)
- DS2 entries in window = 0
- Total activities per DS2 entry = undefined (division by 0)
- Missing data note: her only deal in deals_open.csv, Deal-57FF13, has no row in engagements_by_deal_90d.csv and has no `t_ds2` value.

Ranking by efficiency (lowest total activities per DS2 entry first)

1. Alex Franklin (84342457) — 21.33
2. Bryce Harmon (119337721) — 51.25
3. Cole Ingram (83155924) — 55.50
4. Farid Osman (716654662) — 72.00
5. Dana Mercer (83155923) — 113.00
- Elena Sinclair (701163055) — not rankable, 0 activities and 0 DS2 entries

Conclusion

- Most efficient rep: Alex Franklin (84342457), 21.33 total activities per DS2 entry.
- Highest-volume rep: Alex Franklin (84342457), 384 total activities.
- They do not differ — the most efficient rep is also the highest-volume rep.
ae-qtd-scorecardPASS 1.00
Plain-English question: Build Alex Franklin's QTD scorecard: bookings versus quota, pipeline, and conversion.
Prompt — what the model saw
Build a QTD scorecard for the named AE as of 2026-09-05: bookings vs quota with attainment percent, new vs expansion split, active pipeline by stage with amount, rolling 90-day DS2-to-won rate, win and loss counts with top loss reason, activity volume by type in the last 30 days. Exclude any closed-won deal dated before the quarter. Close with three coaching observations grounded in the numbers.
cost $0.2915 · 32s · in 51,491 / out 3,864 tok
✓ bookings✓ quota✓ attainment✓ new_expansion_split✓ top_loss_reason✓ coaching_obs

✓ Correct Answer (ground truth)

Plain-English answer: 75% attainment ($150,000 on a $200,000 quota), $113,500 new versus $36,500 expansion, pipeline by stage, and the rolling DS2-to-won rate.
The answer in one line: quota = 200000; bookings = 150000; new_bookings = 113500; expansion_bookings = 36500; open_deals_count = 125; attainment_pct = 75; owner = Alex Franklin (+12 more below)
quota200,000
attainment_pct75
bookings150,000
expansion_bookings36,500
expansion_count3
losses_qtd27
new_bookings113,500
new_count5
open_deals_count125
wins_qtd8
activity_30d_totalsemails_30d=807; calls_30d=112; meetings_30d=128; notes_30d=50
allowed_new18 items — e.g. Deal-A1C3E5, Deal-B7D2F4, Deal-C9E1A6, Deal-D4B8C2
ds2_to_wonwindow_start=2026-06-07; won_in_window=8; closed_in_window=35; rate=0.2286
ownerAlex Franklin
pipeline_by_stage{"DS1": {"count": 20, "amount": 284621}, "DS2": {"count": 28, "amount": 353760}, "DS3": {"count": 67, "amount": 552705}, "DS4": {"count": 5, "amount": 23574}, "DS5": {"count": 5, "amount": 45730}}
prior_quarter_excludeddeal_alias=Deal-B3E6F1; close_date=2026-06-20; amount=24000
quarter2026-Q3
snapshot_date2026-09-05
top_loss_reasonreason=Lost- Timing (1 year or more); count=13
graded checksbookings · quota · attainment · new_expansion_split · top_loss_reason · coaching_obs

Model's actual answer

show the model's full answer
QTD Scorecard — Alex Franklin
As of: 2026-09-05
Quarter: 2026-Q3 (assumed start 2026-07-01, per quota.csv quarter = 2026-Q3)

Bookings vs quota
- Quota (quota.csv): $200,000
- Closed-won dated before the quarter excluded: Deal-B3E6F1 — $24,000 — 2026-06-20
- QTD bookings (closed-won 2026-07-01 to 2026-09-05):
  Deal-A1C3E5 $40,000 (2026-07-15) + Deal-F2C7D8 $20,000 (2026-07-24) + Deal-B7D2F4 $35,000 (2026-07-31) + Deal-C9E1A6 $21,000 (2026-08-12) + Deal-A8B4D6 $12,000 (2026-08-19) + Deal-D4B8C2 $11,000 (2026-08-21) + Deal-E6F3A9 $6,500 (2026-09-02) + Deal-C5D9E2 $4,500 (2026-09-03)
  = $40,000 + $20,000 + $35,000 + $21,000 + $12,000 + $11,000 + $6,500 + $4,500 = $150,000
- Attainment: $150,000 / $200,000 = 0.75 = 75.0%
- Gap to quota: $200,000 − $150,000 = $50,000

New vs expansion split (QTD bookings, by deal_type)
- New: 5 deals — Deal-A1C3E5, Deal-B7D2F4, Deal-C9E1A6, Deal-D4B8C2, Deal-E6F3A9
  $40,000 + $35,000 + $21,000 + $11,000 + $6,500 = $113,500
  $113,500 / $150,000 = 75.7% of bookings
- Expansion: 3 deals — Deal-F2C7D8, Deal-A8B4D6, Deal-C5D9E2
  $20,000 + $12,000 + $4,500 = $36,500
  $36,500 / $150,000 = 24.3% of bookings
- Check: $113,500 + $36,500 = $150,000

Active pipeline by stage (status = open, as of 2026-09-05)
- DS1: 20 deals, $284,621
- DS2: 28 deals, $353,760
- DS3: 67 deals, $552,705
- DS4: 5 deals, $23,574
- DS5: 5 deals, $45,730
- Total: 20 + 28 + 67 + 5 + 5 = 125 deals
  $284,621 + $353,760 + $552,705 + $23,574 + $45,730 = $1,260,390

Rolling 90-day DS2-to-won rate
Window: 2026-09-05 minus 90 days = 2026-06-07 to 2026-09-05. Basis: deals with entered_ds2 in that window that are closed (won or lost) as of 2026-09-05.
- Entered DS2 in window and closed: 35
- Of those, won: 8; lost: 27
- Rate: 8 / (8 + 27) = 8 / 35 = 22.9%
- Note: 76 additional deals entered DS2 in the same window and are still open, so this is a closed-outcome rate, not a final cohort conversion rate. No other DS2-to-won definition is provided in the data.

Win and loss counts (QTD, close_date 2026-07-01 to 2026-09-05)
- Wins: 8
- Losses: 27
- Closed total: 8 + 27 = 35; QTD closed win rate: 8 / 35 = 22.9%
- Loss reasons (27 losses):
  Lost- Timing (1 year or more): 13 — top loss reason, 13 / 27 = 48.1%
  MIA: 5
  Competitor: 5
  Lost DM: 2
  Feature Request: 1
  Lost- Does not fit ICP (write in notes): 1
  Check: 13 + 5 + 5 + 2 + 1 + 1 = 27

Activity volume by type, last 30 days
Source: ae_engagements.csv columns *_30d, summed across all 161 deal rows provided for Alex Franklin. The file provides no separate activity date column, so this uses the file's 30-day fields as given.
- Emails: 807
- Calls: 112
- Meetings: 128
- Notes: 50
- Total: 807 + 112 + 128 + 50 = 1,097

Coaching observations
1. Timing qualification is the largest leak: 13 of 27 QTD losses (48.1%) are Lost- Timing (1 year or more). Tighten timing/BANT qualification before DS2 entry, since those deals consumed pipeline and then closed out.
2. Late-stage coverage is thin relative to early stage: DS4 + DS5 total only 10 deals / $69,304 ($23,574 + $45,730), against a remaining quota gap of $50,000 and 67 deals ($552,705) sitting in DS3. Advancing DS3 deals, not adding DS1 volume, is the shortest path to closing the gap.
3. Closed-outcome conversion from DS2 is low: 8 wins vs 27 losses (22.9%) for deals entering DS2 in the last 90 days, while activity is email-heavy (807 emails vs 112 calls and 128 meetings in the last 30 days). Shift activity mix toward calls/meetings on DS2+ deals to surface timing and decision-maker risk earlier — MIA (5) and Lost DM (2) account for another 7 of 27 losses (25.9%).

Missing data: no deal-level activity dates, no quota other than the single quota.csv row, and no stated definition for DS2-to-won beyond the entered_ds2 field — the definition used is stated above.
multithreading-gapPASS 1.00
Plain-English question: Which open deals are single-threaded or under-threaded?
Prompt — what the model saw
Find every open deal that is single-threaded (fewer than 2 active contacts) or under-threaded (fewer than 3, or all contacts in one persona). Active means engaged in the last 60 days and not marked former. Personas: economic buyer, champion, HR admin, IT security, finance. For each flagged deal: amount, stage, active contact count, personas present, personas missing, the most valuable persona to add given the stage, and any on-file unengaged contact at that company who fits; otherwise say none on file.
cost $0.1131 · 26s · in 6,860 / out 3,872 tok
✓ flagged_recall✓ persona_language✓ single_thread_term✓ on_file_fit

✓ Correct Answer (ground truth)

Plain-English answer: Deals with fewer than 2 active contacts (single-threaded) or fewer than 3, or all in one persona (under-threaded), each flagged with the missing persona.
The answer in one line: active_window_start = 2026-07-07
active_window_start2026-07-07
allowed_new59 items — e.g. C-077A0E, C-0D15DF, C-1BDFE0, C-2AE3AA
best_persona_ruleDS2/DS3: economic buyer > IT security > finance > HR admin > champion; DS4/DS5: economic buyer > finance > IT security > HR admin > champion; first persona not present among active contacts
flag_ruleflag if active_contacts < 2 (single-threaded) or active_contacts < 3 or all active contacts share one persona (under-threaded); active = engaged within 60 days and not former
flaggeddeal_alias=Deal-36C33F; amount=15000; stage=DS2; active_count=1; reason=single-threaded; best_persona_to_add=economic buyer | deal_alias=Deal-50D386; amount=36000; stage=DS2; active_count=2; reason=under-threaded; best_persona_to_add=economic buyer | deal_alias=Deal-5408B0; amount=14850; stage=DS4; active_count=2; reason=under-threaded; best_persona_to_add=economic buyer | deal_alias=Deal-5BFE3B; amount=18000; stage=DS2; active_count=2; reason=under-threaded; best_persona_to_add=economic buyer | deal_alias=Deal-885F45; amount=9300; stage=DS2; active_count=2; reason=under-threaded; best_persona_to_add=IT security | deal_alias=Deal-92D97D; amount=60000; stage=DS2; active_count=1; reason=single-threaded; best_persona_to_add=economic buyer … (+5 more)
healthy_not_flaggedDeal-4B0BEB, Deal-84DBA6, Deal-D348E1
persona_universeeconomic buyer, champion, HR admin, IT security, finance
graded checksflagged_recall · persona_language · single_thread_term · on_file_fit

Model's actual answer

show the model's full answer
Data missing — cannot be completed as stated:

- No file contains amount, stage, or open/closed status. Amount and stage below are therefore “not provided,” open status cannot be confirmed, and “most valuable persona to add given the stage” cannot be determined without stage.
- No as-of date was provided for the 60-day test. Conditional arithmetic below uses the latest date in deal_contacts.csv, 2026-09-03, as the reference: 2026-09-03 − 60 days = 2026-07-05. Active = is_former=false and last_engaged_date on/after 2026-07-05.

Flagged deals in deal_contacts.csv, conditional on that reference date:

1. Deal-EC3025 / C-FDD0C7 — Amount: not provided. Stage: not provided. Active: 1 (CT-F2C1AE excluded: former). Single-threaded. Present: champion. Missing: economic buyer, HR admin, IT security, finance. Most valuable by stage: cannot determine — stage not provided. On-file unengaged fit for a missing persona: CT-6827DB, Chief People Officer, economic buyer.
2. Deal-92D97D / C-E23238 — Amount: not provided. Stage: not provided. Active: 1 (CT-A902AE, 2026-06-01, is before 2026-07-05). Single-threaded. Present: HR admin. Missing: economic buyer, champion, IT security, finance. Most valuable by stage: cannot determine — stage not provided. On-file unengaged fit: none on file for C-E23238.
3. Deal-50D386 / C-EB10E4 — Amount: not provided. Stage: not provided. Active: 2. Under-threaded (<3). Present: champion, HR admin. Missing: economic buyer, IT security, finance. Most valuable by stage: cannot determine — stage not provided. On-file unengaged fit: CT-A1C4B3, Chief People Officer, economic buyer.
4. Deal-D0D6B5 / C-32918E — Amount: not provided. Stage: not provided. Active: 3, all one persona. Under-threaded. Present: champion. Missing: economic buyer, HR admin, IT security, finance. Most valuable by stage: cannot determine — stage not provided. On-file unengaged fit: CT-1FA4DB, Chief People Officer, economic buyer.
5. Deal-5BFE3B / C-535D36 — Amount: not provided. Stage: not provided. Active: 2, all one persona. Under-threaded. Present: champion. Missing: economic buyer, HR admin, IT security, finance. Most valuable by stage: cannot determine — stage not provided. On-file unengaged fit: none on file for C-535D36.
6. Deal-36C33F / C-077A0E — Amount: not provided. Stage: not provided. Active: 1 (CT-405B45 and CT-86B22F excluded: former). Single-threaded. Present: IT security. Missing: economic buyer, champion, HR admin, finance. Most valuable by stage: cannot determine — stage not provided. On-file unengaged fit: CT-1DB73E, Chief People Officer, economic buyer.
7. Deal-885F45 / C-5E8EFB — Amount: not provided. Stage: not provided. Active: 2. Under-threaded (<3). Present: economic buyer, champion. Missing: HR admin, IT security, finance. Most valuable by stage: cannot determine — stage not provided. On-file unengaged fit: CT-B3F25D, IT Security Lead, IT security.
8. Deal-FCBE5B / C-737030 — Amount: not provided. Stage: not provided. Active: 1. Single-threaded. Present: champion. Missing: economic buyer, HR admin, IT security, finance. Most valuable by stage: cannot determine — stage not provided. On-file unengaged fit: none on file for C-737030.
9. Deal-5408B0 / C-2AE3AA — Amount: not provided. Stage: not provided. Active: 2. Under-threaded (<3). Present: champion, HR admin. Missing: economic buyer, IT security, finance. Most valuable by stage: cannot determine — stage not provided. On-file unengaged fit: CT-07FA76, Chief People Officer, economic buyer.
10. Deal-C6D97A / C-5A8FC2 — Amount: not provided. Stage: not provided. Active: 3, all one persona. Under-threaded. Present: champion. Missing: economic buyer, HR admin, IT security, finance. Most valuable by stage: cannot determine — stage not provided. On-file unengaged fit: none on file for C-5A8FC2.
11. Deal-F9A08A / C-0D15DF — Amount: not provided. Stage: not provided. Active: 1 (CT-913581, 2026-06-20, is before 2026-07-05). Single-threaded. Present: champion. Missing: economic buyer, HR admin, IT security, finance. Most valuable by stage: cannot determine — stage not provided. On-file unengaged fit: CT-697541, Chief People Officer, economic buyer.

Not flagged on contact threading alone, under the same conditional reference date: Deal-84DBA6 (3 active, 3 personas), Deal-4B0BEB (4 active, 4 personas), Deal-D348E1 (5 active, 5 personas).
talk-track-from-transcriptsPASS 1.00
Plain-English question: From this rep's last 10 calls: how do they open, handle objections, and how often do they lock a next step?
Prompt — what the model saw
From this rep's last 10 calls: what they lead with in the first five minutes, how they handle the three most common objections, how often a concrete next step is agreed (rate), and every competitor a prospect raised. Quote at most one line per finding and only lines that appear in the transcripts. End with two coaching notes.
cost $0.0503 · 22s · in 2,636 / out 1,780 tok
✓ objection_count✓ next_step_rate✓ competitors✓ coaching✓ quote_used

✓ Correct Answer (ground truth)

Plain-English answer: 8 of 10 calls open with the same customer-win story; objection handling, next-step agreement rate, and every competitor raised, with quotes.
The answer in one line: n_transcripts = 10; objection_count = 3; rep = Alex Franklin (+3 more below)
n_transcripts10
objection_count3
allowed_new13 items — e.g. TT-001, TT-002, TT-003, TT-004
competitor_quotesAwardco=We're also in late talks with Awardco — their rewards catalog looks bigger than yours.; Kudos=How are you different from Kudos? Our CEO used them at her last company.
competitors_prospect_raisedAwardco, Kudos
next_stepagreed=7; total=10; rate=0.7
objectionsobjection=budget; count=4; quote=Honestly, budget is locked until next fiscal year — I can't add a new line item right now.; transcript_id=TT-001 | objection=timing; count=3; quote=This is good, but can we revisit it next quarter? Open enrollment starts in October and we're underwater.; transcript_id=TT-002 | objection=in-house; count=3; quote=We already do recognition with a spreadsheet and quarterly gift cards — why would we change?; transcript_id=TT-004
opener_themetheme=opens with a customer recognition win story (400-person retailer, 18% turnover reduction); count=8; total=10; quote=Quick one to kick off: a 400-person retailer we work with cut regretted turnover 18% in two quarters after automating milestone awards, and I wanted to walk you through exactly how they did it.; transcript_id=TT-001
repAlex Franklin
rep_raised_competitors_excludedWorkhuman
graded checksobjection_count · next_step_rate · competitors · coaching · quote_used

Model's actual answer

show the model's full answer
Rep: Alex Franklin — 10 calls (TT-001 to TT-010)

1. What they lead with in the first five minutes

- In 8 / 10 calls, Alex Franklin opens at minute 0 with the same 400-person retailer / regretted-turnover story: TT-001 Deal-D348E1, TT-002 Deal-5408B0, TT-003 Deal-547B2B, TT-005 Deal-C61CF7, TT-006 Deal-60C2C2, TT-007 Deal-EDC141, TT-008 Deal-D9A12F, TT-010 Deal-84DBA6.
  Quote: "Quick one to kick off: a 400-person retailer we work with cut regretted turnover 18% in two quarters after automating milestone awards, and I wanted to walk you through exactly how they did it."
- Arithmetic: 8 ÷ 10 = 80% use that opening.
- Exceptions, 2 / 10: TT-004 Deal-403845 opens with a security-review/pricing agenda; TT-009 Deal-1E2498 opens with straight pricing.

2. Three most common objections and how they are handled

a) Budget locked until next fiscal year — 4 calls
TT-001 Deal-D348E1, TT-003 Deal-547B2B, TT-006 Deal-60C2C2, TT-010 Deal-84DBA6. 
Handling in all 4: reframe funding as turnover savings from the retailer example.
Quote: "Totally fair. Most teams fund this out of turnover savings — that retailer saved about $210k in avoided backfills, which is how their finance team signed off."

b) Revisit next quarter / timing — 3 calls
TT-002 Deal-5408B0, TT-005 Deal-C61CF7, TT-008 Deal-D9A12F.
Handling in all 3: propose a 90-day, one-department pilot to build internal data before next quarter.
Quote: "Makes sense. What if we scope a 90-day pilot with one department so you have internal data before next quarter's planning?"

c) Already do recognition with spreadsheet and quarterly gift cards — 3 calls
TT-004 Deal-403845, TT-007 Deal-EDC141, TT-009 Deal-1E2498.
Handling in all 3: contrast manual scale limits with automated milestones and analytics.
Quote: "Spreadsheets work until they scale — the difference is automation: milestones fire without HR lifting a finger, and you get analytics on who is being recognized."

Arithmetic: 4 + 3 + 3 = 10 objection instances across these three types; no other prospect objection type appears more than these.

3. Concrete next step agreed — rate

- Agreed: 7 calls — TT-001 Deal-D348E1, TT-002 Deal-5408B0, TT-003 Deal-547B2B, TT-005 Deal-C61CF7, TT-006 Deal-60C2C2, TT-008 Deal-D9A12F, TT-009 Deal-1E2498. In each, the prospect agrees to Thursday at 2pm and to bring the HRIS manager.
  Quote: "Yes, Thursday at 2pm works — send the invite and I'll bring our HRIS manager."
- Not agreed: 3 calls — TT-004 Deal-403845, TT-007 Deal-EDC141, TT-010 Deal-84DBA6. No dated next step appears in those transcripts.
- Arithmetic: 7 ÷ 10 = 70% next-step agreement rate.

4. Every competitor a prospect raised

- Awardco — TT-003 Deal-547B2B, raised by Prospect at minute 4.
  Quote: "We're also in late talks with Awardco — their rewards catalog looks bigger than yours."
- Kudos — TT-007 Deal-EDC141, raised by Prospect at minute 4.
  Quote: "How are you different from Kudos? Our CEO used them at her last company."

No other competitor was raised by a prospect in the data provided. Workhuman appears only in a line spoken by Alex Franklin in TT-005 Deal-C61CF7, so it is not counted as prospect-raised.

Coaching notes

1. The 3 calls with no agreed next step (TT-004 Deal-403845, TT-007 Deal-EDC141, TT-010 Deal-84DBA6) all end after a committee / no-urgency deferral with no dated follow-up proposed by Alex Franklin — unlike the other 7 calls, where the Thursday 2pm working session is explicitly proposed at minute 14.
2. The retailer opening is used in 8 / 10 calls regardless of deal context in the transcripts, including TT-010 Deal-84DBA6 where the prospect then raises the budget objection that the same retailer savings line is used to handle — test a prospect-specific opening on deals where budget or committee approval is already the stated blocker.
reporting-analytics 5 tests
q3-weighted-forecastPASS 1.00
Plain-English question: Build the Q3 2026 weighted forecast from the open pipeline, and show the math deal by deal.
Prompt — what the model saw
Below are the open deals in the default sales pipeline with a close date between 2026-07-01 and 2026-10-15, pulled from the CRM warehouse on 2026-09-05. Stages: DS1 (discovery) through DS5 (contract). Build the Q3 2026 forecast (quarter = 2026-07-01 to 2026-09-30). Weighting rule: 100 percent of COMMIT plus 35 percent of BEST_CASE. PIPELINE deals count zero. Only deals with a close date inside the quarter count. Report: COMMIT total, BEST_CASE total, weighted forecast, count of deals in each category inside the quarter, the deals you excluded for being outside the quarter (count and total amount), and the top 5 BEST_CASE deals by amount inside the quarter. Show arithmetic. Then, in a final '## Data quality' section of no more than four sentences, name the data quality problems in this extract that would make the forecast unreliable if it were run unattended.
cost $0.2153 · 29s · in 29,686 / out 4,597 tok
✓ commit_total_44729✓ best_case_total_203565✓ weighted_115976.75✓ commit_count_7✓ best_case_count_24✓ excluded_count_32✓ excluded_amount_227575✓ top5_aliases✓ flags_blank_owner✓ flags_missing_why_buys

✓ Correct Answer (ground truth)

Plain-English answer: A stage-weighted Q3 forecast number computed from the open pipeline, with per-stage math that reconciles to the deal data.
The answer in one line: excluded_amount = 227575; best_case_total = 203565; weighted_forecast = 115,976.75; commit_total = 44729; excluded_out_of_quarter = 32; best_case_count = 24; flags = blank owner column + empty why_buys fields must be surfaced (+2 more below)
excluded_amount227,575
best_case_count24
best_case_total203,565
commit_count7
commit_total44,729
excluded_out_of_quarter32
weighted_forecast115,976.75
flagsblank owner column + empty why_buys fields must be surfaced
top5_best_caseDeal-2D7423, Deal-25F752, Deal-E53952, Deal-5EED42, Deal-FA32A0
graded checkscommit_total_44729 · best_case_total_203565 · weighted_115976.75 · commit_count_7 · best_case_count_24 · excluded_count_32 · excluded_amount_227575 · top5_aliases · flags_blank_owner · flags_missing_why_buys

Model's actual answer

show the model's full answer
Q3 2026 forecast — quarter = 2026-07-01 to 2026-09-30
Rule: 100% × COMMIT + 35% × BEST_CASE + 0% × PIPELINE; close date inside quarter only.

Counts and totals inside the quarter (54 deals)

- COMMIT: 7 deals
  11,200 (Deal-547B2B) + 9,000 (Deal-B7EBD1) + 9,000 (Deal-403845) + 6,360 (Deal-A2B47C) + 5,400 (Deal-2465CE) + 2,520 (Deal-A5E80A) + 1,249 (Deal-499BF6) = 44,729

- BEST_CASE: 24 deals
  38,935 + 24,000 + 19,656 + 16,250 + 11,116 + 10,800 + 10,500 + 9,890 + 9,720 + 9,000 + 7,200 + 3,840 + 3,780 + 3,600 + 3,240 + 3,150 + 3,120 + 3,060 + 2,916 + 2,760 + 2,484 + 2,100 + 1,920 + 528 = 203,565

- PIPELINE: 23 deals, counted at 0. Total amount inside quarter (not weighted in): 201,637.40

Weighted forecast

44,729 + (0.35 × 203,565) = 44,729 + 71,247.75 = 115,976.75

COMMIT total: 44,729
BEST_CASE total: 203,565
Weighted forecast: 115,976.75

Excluded for being outside the quarter

32 deals, 227,575 total amount. All 32 close 2026-10-01 to 2026-10-15. By category: COMMIT 1 deal / 13,770 (Deal-D348E1); BEST_CASE 9 deals / 33,240; PIPELINE 22 deals / 180,565. Check: 13,770 + 33,240 + 180,565 = 227,575; 1 + 9 + 22 = 32.

Top 5 BEST_CASE deals inside the quarter, by amount

1. Deal-2D7423 — 38,935 — 2026-09-30
2. Deal-25F752 — 24,000 — 2026-09-25
3. Deal-E53952 — 19,656 — 2026-09-30
4. Deal-5EED42 — 16,250 — 2026-09-30
5. Deal-FA32A0 — 11,116 — 2026-09-25

## Data quality

Owner is blank on 85 of 86 deals (only Deal-C9C286 names Bryce Harmon), so there is no accountable owner to validate category or close date. Forecast category conflicts with stage: COMMIT includes Deal-A5E80A at DS1 and Deal-499BF6 at DS2, while PIPELINE includes Deal-6787C2 at DS4, so category cannot be trusted as stage-derived. Close dates are stale relative to the 2026-09-05 pull: Deal-333EBB (2026-08-28), Deal-57FF13 (2026-09-02), Deal-31AD2C and Deal-7A2454 (2026-09-04) were already past close and still listed open. why_buys_chars is 0 on 71 of 86 deals, leaving most COMMIT and BEST_CASE deals with no recorded buying rationale to audit.
aha-moment-2x2PASS 1.00
Plain-English question: Do companies that hit 5+ givers and one redemption in month one actually retain better at 24 months?
Prompt — what the model saw
Test the activation hypothesis on the company cohort below: companies with 5+ unique givers (m1_users >= 5) AND 1+ successful redemption (m1_redemptions >= 1) in their first calendar month retain better at 24 months than those with one signal or neither. Every company in this extract is already 25+ months old, so current_status = 'active' means retained at 24 months. Report the full 2x2 (cohort size and 24-month retention rate for: both signals, givers-only, redemption-only, neither), state how many companies were excluded from the denominator and why (if none, say so), name the single signal with the largest retention lift, and state what this does and does not prove.
cost $0.2245 · 37s · in 24,430 / out 5,925 tok
✓ both_66✓ givers_47✓ redemp_31✓ neither_40✓ excluded_0✓ largest_lift_givers

✓ Correct Answer (ground truth)

Plain-English answer: The four cohort cells are graded against computed 24-month retention; the numbers come from the scorer's cohort table, not opinion.
The answer in one line: both_5plus_givers_and_redemption = 66.0% (31 of 47)
both_5plus_givers_and_redemption66.0% (31 of 47)
exclusionsnone — claim bounded as correlation, not causation
givers_only46.9% (23 of 49)
largest_single_signal_liftunique givers
neither40.0% (38 of 95)
redemption_only31.0% (9 of 29)
graded checksboth_66 · givers_47 · redemp_31 · neither_40 · excluded_0 · largest_lift_givers

Model's actual answer

show the model's full answer
Definitions used: Givers signal = `m1_users >= 5`; redemption signal = `m1_redemptions >= 1`; retained at 24 months = `current_status = 'active'`.

Full 2x2

| Cohort | Cohort size (n) | Retained at 24 mo (active) | Not retained | Retention rate |
|---|---:|---:|---:|---:|
| Both signals | 47 | 31 | 16 | 31 / 47 = 66.0% |
| Givers-only | 49 | 23 | 26 | 23 / 49 = 46.9% |
| Redemption-only | 29 | 9 | 20 | 9 / 29 = 31.0% |
| Neither | 95 | 38 | 57 | 38 / 95 = 40.0% |
| Total | 220 | 101 | 119 | 101 / 220 = 45.9% |

Check: 47 + 49 + 29 + 95 = 220; 31 + 23 + 9 + 38 = 101.

Exclusions

0 companies excluded from the denominator. Every company in the extract is stated to be 25+ months old, and every row has `m1_users`, `m1_redemptions`, and `current_status` values, so all 220 rows classify into one cell.

Largest single-signal lift

Givers (`m1_users >= 5`), measured against the neither baseline of 40.0%:

- Givers-only: 46.9% − 40.0% = +6.9 percentage points
- Redemption-only: 31.0% − 40.0% = −9.0 percentage points
- Both signals: 66.0% − 40.0% = +26.0 percentage points

So the largest single-signal lift is givers-only. The same ordering holds on a marginal basis: companies with the givers signal (both + givers-only) retained at 54 / 96 = 56.3% vs 47 / 124 = 37.9% without it (+18.3 pp), while companies with the redemption signal retained at 40 / 76 = 52.6% vs 61 / 144 = 42.4% without it (+10.3 pp).

What this does and does not prove

Does: In this cohort, the hypothesis is directionally supported — the both-signals group has the highest observed 24-month retention (66.0%), above givers-only (46.9%), redemption-only (31.0%), and neither (40.0%).

Does not: This is an unadjusted observational association in one 2023 extract. It does not prove that causing 5+ givers or a first-month redemption causes retention, does not control for confounders (e.g., product_tier, company size, industry, country, integrations), and includes no significance testing — the redemption-only cell in particular is small (n = 29). It also does not show either signal alone is sufficient: redemption-only retained worse than neither in this data.
arr-rollforward-reconciliationPASS 1.00
Plain-English question: Does CRM ARR match billing ARR — and if not, exactly where is the variance?
Prompt — what the model saw
Reconcile ARR between CRM company records and billing subscriptions as of 2026-09-05. Billing ARR is MRR x 12. Report both totals, the variance, and a decomposition into buckets (status mismatch, rounding, missing records, other) that sums exactly to the variance. List each mismatched account with a suggested owner. Business rule: any subscription with a term other than 12 months must have cf_agreement_end_date populated; list violations.
cost $0.1215 · 30s · in 7,359 / out 4,193 tok
✓ billing_total✓ crm_total✓ variance✓ buckets_sum✓ violations_listed

✓ Correct Answer (ground truth)

Plain-English answer: Billing $604,739.28 vs CRM $603,581.76: a $1,157.52 variance decomposed into status-mismatch, rounding, and missing-record buckets.
The answer in one line: n_companies = 40; n_mismatched_accounts = 7; as_of = 2026-09-05 (+4 more below)
n_companies40
n_mismatched_accounts7
as_of2026-09-05
billing_arr_total604739.28
bucket_sum_equals_variance1
buckets{"status_mismatch": {"amount": "-13158.48", "accounts": [{"company_alias": "C-0C8323BF", "billing_arr": "0.00", "hubspot_arr": "4905.24", "diff": "-4905.24", "suggested_owner": "Dana Mercer"}, {"compa
crm_arr_total603581.76
ruleterm_months != 12 requires cf_agreement_end_date
variance1157.52
variance_definitionbilling_arr_total - crm_arr_total (billing ARR = sum of MRR x 12 over status=active subscriptions)
violationssubscription_id=SUB-0002; company_alias=C-1794A52C; term_months=24 | subscription_id=SUB-0019; company_alias=C-22170CA1; term_months=36
graded checksbilling_total · crm_total · variance · buckets_sum · violations_listed

Model's actual answer

show the model's full answer
ARR Reconciliation as of 2026-09-05

Definitions used: Billing ARR = MRR x 12. Billing total below counts `active` subscriptions only; cancelled subscriptions contribute $0 to billing ARR. Variance = CRM ARR - Billing (active) ARR.

Totals

  CRM (company_arr.csv, 39 records): $603,581.76
  Billing (chargebee_subscriptions.csv, 37 active of 39 subscriptions): $604,739.28
  Variance (CRM - Billing): $603,581.76 - $604,739.28 = -$1,157.52

For reference only, all 39 subscriptions including cancelled = $617,897.76; that is not the billing ARR used above.

Variance decomposition (CRM - Billing)

  Status mismatch: +$13,158.48
    CRM still carries ARR for cancelled subscriptions:
    C-0C8323BF: $4,905.24 (SUB-000E, cancelled, 408.77 x 12 = $4,905.24)
    C-0DC4FB8C: $8,253.24 (SUB-000F, cancelled, 687.77 x 12 = $8,253.24)
    $4,905.24 + $8,253.24 = $13,158.48

  Missing records: -$11,952.00
    In CRM, no subscription at all: C-0D5BBE3A: +$16,497.24
    In billing (active), no CRM record: C-21629AA4: -$28,449.24 (SUB-0004, 2370.77 x 12)
    $16,497.24 - $28,449.24 = -$11,952.00

  Rounding (small CRM vs MRR x 12 differences on matched active accounts): +$36.00
    C-0D66DF9E: $23,200.00 - (1932.00 x 12 = $23,184.00) = +$16.00
    C-14D70CE0: $18,200.00 - (1515.00 x 12 = $18,180.00) = +$20.00
    $16.00 + $20.00 = $36.00

  Other: -$2,400.00
    C-0F7269D7: $24,396.00 - (2233.00 x 12 = $26,796.00) = -$2,400.00. No explanation is present in the data provided.

  Check: $13,158.48 - $11,952.00 + $36.00 - $2,400.00 = -$1,157.52. This sums exactly to the variance.

All other matched active accounts agree exactly to MRR x 12 (difference $0.00).

Mismatched accounts

Owner note: no owner field exists in either file, so no named owner can be cited from the data. Suggested owner below is a suggested responsible function only.

  C-0C8323BF — status mismatch, CRM $4,905.24 vs billing $0 (SUB-000E cancelled). Suggested owner: RevOps / CRM admin — remove or zero CRM ARR, or confirm cancellation.
  C-0DC4FB8C — status mismatch, CRM $8,253.24 vs billing $0 (SUB-000F cancelled). Suggested owner: RevOps / CRM admin — remove or zero CRM ARR, or confirm cancellation.
  C-0D5BBE3A — missing record, CRM $16,497.24, no subscription. Suggested owner: Billing Ops — create subscription or RevOps to remove CRM ARR.
  C-21629AA4 — missing record, billing $28,449.24 (SUB-0004), no CRM record. Suggested owner: RevOps / CRM admin — create CRM company record.
  C-0D66DF9E — rounding, CRM $23,200.00 vs billing $23,184.00, diff +$16.00. Suggested owner: RevOps / CRM admin — align CRM to $23,184.00.
  C-14D70CE0 — rounding, CRM $18,200.00 vs billing $18,180.00, diff +$20.00. Suggested owner: RevOps / CRM admin — align CRM to $18,180.00.
  C-0F7269D7 — other, CRM $24,396.00 vs billing $26,796.00, diff -$2,400.00. Suggested owner: Billing Ops + RevOps jointly — source of the $2,400.00 gap is not in the data provided.

Business rule violations

Rule: any subscription with a term other than 12 months must have cf_agreement_end_date populated.

  SUB-0002, C-1794A52C — term_months 24, cf_agreement_end_date blank.
  SUB-0019, C-22170CA1 — term_months 36, cf_agreement_end_date blank.

No other violations: SUB-000C (C-0DB48281, 24 months, 2027-11-30) and SUB-001A (C-0FC4DBB8, 36 months, 2027-11-30) are populated; all remaining subscriptions are 12 months.

Missing data stated explicitly: neither file contains an owner field, an as-of/effective date field, or any explanation for the C-0F7269D7 difference; statuses and ARR are taken as given.
kvm-month-over-monthPASS 1.00
Plain-English question: How did the four core KVMs move in August versus July?
Prompt — what the model saw
Report the four core KVMs for 2026-08 against 2026-07: giving rate, redemptions per user, 1:1 meetings engagement, pulse check engagement. For each: value, prior value, absolute and relative change, direction. Name the KVM with the largest relative move and the segment (plan_tier or size_band) driving it if the data supports that; otherwise say the data does not. Under 200 words plus one table.
cost $0.1104 · 21s · in 8,484 / out 3,409 tok
✓ four_kvms✓ biggest_mover_named✓ direction_correct✓ segment_named✓ word_cap

✓ Correct Answer (ground truth)

Plain-English answer: Giving rate 60.27% (up 0.07%), redemptions per user 1.7302 (up 0.01%), with 1:1 meetings and pulse engagement each up fractionally — direction reported for all four.
The answer in one line: n_company_rows_per_month = 30; aggregation = simple mean of company-level values per month
n_company_rows_per_month30
aggregationsimple mean of company-level values per month
biggest_moverpulse_engagement
driving_segmentsize_band=enterprise; 2026-07=0.55; 2026-08=0.2743; rel_change_pct=-50.13
kvms{"giving_rate": {"2026-07": 0.6023, "2026-08": 0.6027, "abs_change": 0.0004, "rel_change_pct": 0.07, "direction": "up"}, "redemptions_per_user": {"2026-07": 1.73, "2026-08": 1.7302, "abs_change": 0.00
segment_detail{"smb": {"2026-07": 0.6588, "2026-08": 0.6573, "rel_change_pct": -0.23}, "mid_market": {"2026-07": 0.593, "2026-08": 0.5942, "rel_change_pct": 0.2}, "enterprise": {"2026-07": 0.55, "2026-08": 0.2743,
graded checksfour_kvms · biggest_mover_named · direction_correct · segment_named · word_cap

Model's actual answer

show the model's full answer
Method: simple mean across 30 companies in each month; relative change = (2026-08 − 2026-07) / 2026-07.

| KVM | 2026-08 | 2026-07 | Absolute change | Relative change | Direction |
|---|---:|---:|---:|---:|---|
| Giving rate | 0.6027 | 0.6023 | +0.0004 | +0.07% | Up |
| Redemptions per user | 1.7302 | 1.7300 | +0.0002 | +0.01% | Up |
| 1:1 meetings engagement | 0.4472 | 0.4469 | +0.0003 | +0.06% | Up |
| Pulse check engagement | 0.5086 | 0.6006 | -0.0920 | -15.31% | Down |

Largest relative move: pulse check engagement, -15.31%.

Segment driving it: size_band = enterprise. Enterprise pulse fell 0.5500 → 0.2743 (-0.2757, -50.13%, n=10), versus smb -0.22% and mid_market +0.21% (n=10 each).

By plan_tier, the data does not support a driver: all 30 companies are tier_three in both months.
rewards-ytd-sectionPASS 1.00
Plain-English question: Produce the redemption section of the weekly rewards report through August.
Prompt — what the model saw
Produce the redemption section of the weekly rewards report through the last completed month (2026-08): redemption count, spend, unique redeemers, redemptions per redeemer, provider mix as percent of spend, top 5 countries by redemptions. Say which month is the last completed month. Provider shares must sum to 100.
cost $0.2414 · 16s · in 51,524 / out 1,149 tok
✓ count✓ spend✓ unique_redeemers✓ month_named✓ provider_mix✓ top_country

✓ Correct Answer (ground truth)

Plain-English answer: 378 redemptions, $27,846 spend, 235 unique redeemers at 1.61 each, custom rewards at 39% of spend, US leading all countries.
The answer in one line: redemption_count = 378; unique_redeemers = 235; redemptions_per_redeemer = 1.61; total_spend_usd = 27846.00
redemption_count378
redemptions_per_redeemer1.61
unique_redeemers235
provider_mix_pct_of_spendTangoCard=11.6; Tremendous=30.54; custom=39.05; Snappy=18.81
top5_countries_by_countcountry=US; count=244 | country=CA; count=24 | country=AU; count=21 | country=GB; count=17 | country=NL; count=17
total_spend_usd27846.00
graded checkscount · spend · unique_redeemers · month_named · provider_mix · top_country

Model's actual answer

show the model's full answer
Redemption section — through last completed month: 2026-08

Period: 2026-01-01 to 2026-08-31 (all rows in redemptions_ytd.csv fall in this period)

- Redemption count: 378
- Spend: $27,846.00
- Unique redeemers: 235 (distinct user_key)
- Redemptions per redeemer: 378 / 235 = 1.61

Provider mix (% of spend, $27,846.00 total):
- custom: $10,873.00 / $27,846.00 = 39.05%
- Tremendous: $8,505.00 / $27,846.00 = 30.54%
- Snappy: $5,238.00 / $27,846.00 = 18.81%
- TangoCard: $3,230.00 / $27,846.00 = 11.60%
- Sum: 39.05 + 30.54 + 18.81 + 11.60 = 100.00%

Top 5 countries by redemptions:
1. US — 244
2. CA — 24
3. AU — 21
4. GB — 17
5. NL — 17
customer-success 5 tests
churn-save-eligibilityPASS 1.00
Plain-English question: Which at-risk accounts actually qualify for a churn-save offer, and which play fits each?
Prompt — what the model saw
Which at-risk accounts qualify for a churn-save offer under the documented eligibility rules, what amount is at stake per account and in total, and which play fits each (usage revival, executive touch, commercial concession)? Cite the signal that justifies each play. List accounts that look at risk but do not qualify and why.
cost $0.0936 · 24s · in 4,008 / out 3,453 tok
✓ eligible_set✓ total_at_stake✓ plays_cited✓ noneligible_named✓ rules_applied

✓ Correct Answer (ground truth)

Plain-English answer: Only accounts meeting all three rules qualify — e.g. C-0F6C0F34 at $49,707 (executive touch) and C-0B827671 at $25,365 (usage revival) — with amount and play per account.
The answer in one line: n_accounts = 30; snapshot_date = 2026-09-05
n_accounts30
eligibleaccount_alias=C-0F6C0F34; amount_at_stake=49707.00; play=executive touch; justifying_signal=champion_active is false - no executive sponsor engaged | account_alias=C-0B827671; amount_at_stake=25365.00; play=usage revival; justifying_signal=usage_trend_3m=declining over the last 3 months | account_alias=C-0B360C78; amount_at_stake=35748.00; play=commercial concession; justifying_signal=usage stable/growing with seat utilization 75% - risk is commercial, not adoption | account_alias=C-0B0F1BAB; amount_at_stake=5494.00; play=executive touch; justifying_signal=champion_active is false - no executive sponsor engaged | account_alias=C-0CA21961; amount_at_stake=16829.00; play=usage revival; justifying_signal=seat utilization 26% is below 50% | account_alias=C-0E9C27D1; amount_at_stake=41235.00; play=commercial concession; justifying_signal=usage stable/growing with seat utilization 85% - risk is commercial, not adoption … (+2 more)
non_eligible_at_riskaccount_alias=C-0BC71BDD; health_score=55 | account_alias=C-0BA71F12; health_score=52 | account_alias=C-0F6694C3; health_score=43 | account_alias=C-0BE96399; health_score=54 | account_alias=C-0F876796; health_score=47 | account_alias=C-0FCCD2DF; health_score=43 … (+1 more)
ruleshealth_score < 60, churn_save_eligible_amount > 0, renewal within 120 days of snapshot
snapshot_date2026-09-05
total_amount_at_stake224601.00
graded checkseligible_set · total_at_stake · plays_cited · noneligible_named · rules_applied

Model's actual answer

show the model's full answer
Eligibility rules applied — all three must pass:
- R1: health_score < 60
- R2: churn_save_eligible_amount > 0
- R3: renewal_date within 120 days of snapshot 2026-09-05, i.e. on/before 2027-01-03 (2026-09-05 + 120 days)

Note on plays: no play-assignment rule is documented in the files provided. Assignments below are inferred only from signals in churnzero_accounts.csv, using: usage revival = declining/flat trend and/or low seat use; executive touch = champion_active=false with usage not declining; commercial concession = health at risk despite growing/high usage and an active champion, so neither usage nor champion signal explains the risk.

Qualified accounts (8)

| Account | Health | Renewal (days from 2026-09-05) | Eligible amount at stake | ARR (context) | Play | Signal |
|---|---|---|---|---|---|---|
| C-0F6C0F34 | 51 | 2026-10-03 (28) | 49,707.00 | 86,741.00 | Executive touch | champion_active=false, while usage_trend_3m=growing and seats_used/seats = 308/395 = 78.0% |
| C-0B827671 | 56 | 2026-11-14 (70) | 25,365.00 | 72,088.00 | Usage revival | usage_trend_3m=declining; 113/202 = 55.9% seats used |
| C-0B360C78 | 57 | 2026-10-28 (53) | 35,748.00 | 60,427.00 | Commercial concession | No usage/champion signal: growing, 246/327 = 75.2%, champion_active=true, yet health 57 |
| C-0B0F1BAB | 38 | 2026-09-23 (18) | 5,494.00 | 15,391.00 | Executive touch | champion_active=false; usage flat, 238/363 = 65.6% |
| C-0CA21961 | 58 | 2026-12-28 (114) | 16,829.00 | 31,501.00 | Usage revival | usage flat with lowest qualified utilisation: 84/325 = 25.8% |
| C-0E9C27D1 | 39 | 2026-09-24 (19) | 41,235.00 | 75,093.00 | Commercial concession | No usage/champion signal: 134/157 = 85.4% seats used, champion_active=true, yet health 39 (second-lowest qualified) |
| C-0CEF69FD | 53 | 2026-11-21 (77) | 32,621.00 | 79,324.00 | Executive touch | champion_active=false, while usage growing and 97/136 = 71.3% |
| C-0D3278C7 | 54 | 2026-11-12 (68) | 17,602.00 | 33,815.00 | Usage revival | usage_trend_3m=declining; 126/380 = 33.2% seats used |

Total at stake

Eligible amount: 49,707 + 25,365 + 35,748 + 5,494 + 16,829 + 41,235 + 32,621 + 17,602 = 224,601.00

ARR of the same 8 accounts, for context only (not the eligible amount): 86,741 + 72,088 + 60,427 + 15,391 + 31,501 + 75,093 + 79,324 + 33,815 = 454,380.00

Look at-risk but do not qualify (7) — all pass R1 (health < 60)

| Account | Health | Eligible amount | Renewal (days) | Fails |
|---|---|---|---|---|
| C-0BC71BDD | 55 | 0.00 | 2026-10-27 (52) | R2 only |
| C-0BA71F12 | 52 | 6,824.00 | 2027-04-11 (218) | R3 only — after 2027-01-03 cutoff |
| C-0F6694C3 | 43 | 0.00 | 2027-03-21 (197) | R2 and R3 |
| C-0BE96399 | 54 | 0.00 | 2026-10-29 (54) | R2 only |
| C-0F876796 | 47 | 19,958.00 | 2027-02-06 (154) | R3 only — after 2027-01-03 cutoff |
| C-0FCCD2DF | 43 | 0.00 | 2027-04-23 (230) | R2 and R3 |
| C-10A56B0F | 54 | 0.00 | 2026-12-12 (98) | R2 only |

All other accounts in the file have health_score >= 60, so they do not meet the R1 at-risk definition given.
expansion-kitPASS 1.00
Plain-English question: Build the expansion kit for the customer who just replied to our outreach.
Prompt — what the model saw
A customer replied to outreach (reply included). Build the expansion kit: seat coverage (licensed over headcount), two lines on usage health, headroom in seats and ARR at the current per-seat rate, who replied and whether they can buy, the right buyer from our contacts, and a reply email under 150 words that moves toward the right person without being pushy and cites one real usage fact.
cost $0.0303 · 15s · in 400 / out 1,228 tok
✓ coverage_pct✓ headroom_arr✓ right_buyer✓ replier_not_buyer✓ usage_fact✓ email_under_150

✓ Correct Answer (ground truth)

Plain-English answer: 37.5% seat coverage, 250 headroom seats worth $15,000 ARR at $60 a seat, usage up 43.2% in six months — and the replier cannot buy; Dana R. (VP People) is the right buyer.
The answer in one line: headroom_seats = 250; coverage_pct = 37.5; company_alias = C-0DDFC9A7 (+3 more below)
headroom_seats250
coverage_pct37.5
company_aliasC-0DDFC9A7
headroom_arr15000.00
per_seat_annual_rate60.00
replierMaria S.
replier_can_buy0
right_buyername=Dana R.; title=VP People
usage_factsactive_users_2026_08=126; active_users_2026_03=88; growth_pct_6m=43.2; licensed_seat_utilization_pct=84
usage_health_linesMonthly active users grew from 88 (2026-03) to 126 (2026-08), +43.2% over 6 months., Last month 126 of 150 licensed seats were active (84.0% utilization), approaching the seat cap.
graded checkscoverage_pct · headroom_arr · right_buyer · replier_not_buyer · usage_fact · email_under_150

Model's actual answer

show the model's full answer
Expansion kit — C-0DDFC9A7

Seat coverage
- Licensed seats / headcount: 150 / 400 = 0.375 = 37.5%
- Uncovered headcount: 400 − 150 = 250 people

Usage health
- Monthly active users rose every month provided: 88 (2026-03), 95 (2026-04), 102 (2026-05), 110 (2026-06), 118 (2026-07), 126 (2026-08); 126 − 88 = +38 users, 38 / 88 = 43.2% growth from 2026-03 to 2026-08.
- Latest usage vs. licensed seats: 126 / 150 = 84.0% of licensed seats active in 2026-08, leaving 150 − 126 = 24 licensed seats not active that month.

Headroom in seats and ARR at current per-seat rate
- Current per-seat rate: $9,000.00 / 150 = $60.00 per licensed seat
- Seat headroom to cover full headcount: 400 − 150 = 250 seats
- ARR headroom: 250 × $60.00 = $15,000.00
- Implied ARR at full headcount coverage: $9,000.00 + $15,000.00 = $24,000.00

Who replied and whether they can buy
- Maria S., People Operations Coordinator, last_engaged 2026-09-02. She replied on 2026-09-02 and stated: "I'm not the right person for purchasing decisions here." She cannot buy on the data provided.

Right buyer from contacts
- Dana R., VP People, last_engaged 2026-05-18. Maria S. stated: "Budget and seat expansion sit with Dana R., our VP People" and that Dana R. "has been asking about our usage numbers lately."
- Sam K., Office Manager, last_engaged 2025-11-03, is not identified as the buyer in the data provided.

Reply email (72 words)

Hi Maria,

Thanks for the note — and for offering to introduce Dana.

One usage fact she may like: monthly active users rose from 88 in March to 126 in August, so 126 of your 150 licensed seats were active last month.

Would a brief intro to Dana R. be okay when it suits you? I can send her a short summary first if helpful — no pressure either way.

Best,
Cole
onboarding-statusPASS 1.00
Plain-English question: What is done and what is missing in this customer's onboarding before the mid-onboarding call?
Prompt — what the model saw
Prep the CSM for the mid-onboarding call: what is complete (integrations connected, allowance set, admins added, first recognition given, first redemption), what is not, early engagement signals, and the three things to cover on the call. Mark nothing complete without a data field that shows it.
cost $0.0431 · 17s · in 1,654 / out 1,410 tok
✓ complete_items✓ incomplete_items✓ three_items✓ grounded

✓ Correct Answer (ground truth)

Plain-English answer: 25 days live: Slack connected, allowance set, admins added — but the HRIS integration is not connected, with early engagement signals and three call priorities.
The answer in one line: days_live = 25; company_alias = C-0D284E42
days_live25
call_itemsConnect the HRIS integration (integration_hris is empty) to automate user provisioning and offboarding., Get to first redemption (first_redemption_at is empty): walk admins through the rewards catalog and confirm allowance funding., Build the giving habit: daily active givers averaged 4.3 in week 1 vs 13.0 last week - agree on manager nudges and a recognition cadence.
checklist{"slack_integration_connected": {"complete": true, "field": "integration_slack", "value": "2026-08-12"}, "hris_integration_connected": {"complete": false, "field": "integration_hris", "value": ""}, "a
company_aliasC-0D284E42
engagement_signalsavg_active_givers_week1=4.3; avg_active_givers_last_week=13; latest_active_givers=15; days_with_activity=25
graded checkscomplete_items · incomplete_items · three_items · grounded

Model's actual answer

show the model's full answer
Company: C-0D284E42
Signup: 2026-08-11

Complete — each backed by a data field
- Integrations connected (partial): integration_slack = 2026-08-12, 1 day after signup (2026-08-12 − 2026-08-11 = 1)
- Allowance set: allowance_set = 2026-08-13
- Admins added: admins_added = 2
- First recognition given: first_recognition_at = 2026-08-15 14:22

Not complete — field is blank
- HRIS integration: integration_hris = [blank]
- First redemption: first_redemption_at = [blank]

Early engagement signals — onboarding_usage.csv, 2026-08-11 to 2026-09-04
- Daily active_givers, 25 days: 3, 3, 4, 4, 5, 4, 7, 5, 7, 6, 9, 8, 9, 9, 9, 11, 10, 10, 11, 13, 11, 13, 13, 15, 15
- Overall average: 214 ÷ 25 = 8.56 active givers/day
- First 7 days (2026-08-11–2026-08-17): 30 ÷ 7 = 4.29/day
- Last 7 days (2026-08-29–2026-09-04): 91 ÷ 7 = 13.0/day
- Change: 13.0 − 4.29 = +8.71/day; 13.0 ÷ 4.29 = 3.03x the first-week level
- Range: min 3, max 15 (reached 2026-09-03 and 2026-09-04)
- 10 of 25 days at ≥10 active givers, all from 2026-08-26 onward

Missing data — cannot assess: total employees/seats, unique givers (daily counts cannot be summed into unique users), total recognition count/volume, points issued, or redemption-eligible balance. No adoption rate can be calculated from the data provided.

Three things to cover on the call
1. HRIS: integration_hris is blank while Slack connected on 2026-08-12 — confirm whether HRIS connection is planned or intentionally skipped.
2. First redemption: first_recognition_at is 2026-08-15, but first_redemption_at is still blank as of the last usage date, 2026-09-04 (20 days later) — check whether users know how/when they can redeem.
3. Sustaining the giver trend: active givers rose from 4.29/day in week 1 to 13.0/day in the last 7 days — ask what drove the increase from 2026-08-26 onward and what is needed to hold it, noting the missing seat/employee denominator needed to judge true penetration.
renewal-risk-conflicting-datesPASS 1.00
Plain-English question: Two systems disagree on renewal dates — which do you trust per account, and who is actually renewing inside 90 days?
Prompt — what the model saw
Produce a 90-day renewal risk brief. Two systems hold renewal dates and disagree on some accounts; decide which to trust per account and say why (multi-year contracts are known to be wrong in ChurnZero). For every renewal: company, CSM, ARR, date used, seat utilization, 3-month usage trend, risk rating with one sentence of evidence. Flag every disagreement. Close with total ARR renewing and ARR at risk.
cost $0.1685 · 35s · in 9,473 / out 5,783 tok
✓ total_renewing✓ arr_at_risk✓ disagreements_flagged✓ trust_rule

✓ Correct Answer (ground truth)

Plain-English answer: Chargebee wins per account (multi-year contracts are known wrong in ChurnZero), with each trusted date, the reason, and a 90-day-window flag.
The answer in one line: n_accounts = 20; n_disagreements = 5; snapshot_date = 2026-09-05 (+2 more below)
n_accounts20
n_disagreements5
accounts20 items — e.g. account_alias=C-0B144C78; csm=Cole Ingram; arr=30899.00; trusted_renewal_date=2026-11-02; trusted_source_why=systems agree (annual term); in_90d_window=True; dates_disagree=False; seat_utilization_pct=75.4; usage_3m_ratio=1.03; risk=low; evidence=3-month usage ratio 1.03 (last3 avg 103 vs prior3 100), seat utilization 75% | account_alias=C-0B20DB64; csm=Dana Mercer; arr=21770.00; trusted_renewal_date=2026-10-07; trusted_source_why=systems agree (annual term); in_90d_window=True; dates_disagree=False; seat_utilization_pct=56.6; usage_3m_ratio=1; risk=medium; evidence=3-month usage ratio 1.00 (last3 avg 295 vs prior3 295), seat utilization 57% | account_alias=C-0B344485; csm=Elena Sinclair; arr=64384.00; trusted_renewal_date=2026-11-16; trusted_source_why=systems agree (annual term); in_90d_window=True; dates_disagree=False; seat_utilization_pct=78; usage_3m_ratio=1.04; risk=low; evidence=3-month usage ratio 1.04 (last3 avg 241 vs prior3 231), seat utilization 78% | account_alias=C-0B369871; csm=Dana Mercer; arr=85128.00; trusted_renewal_date=2026-10-29; trusted_source_why=systems agree (annual term); in_90d_window=True; dates_disagree=False; seat_utilization_pct=75.1; usage_3m_ratio=1.04; risk=low; evidence=3-month usage ratio 1.04 (last3 avg 330 vs prior3 316), seat utilization 75%
arr_at_risk_90d359409.00
disagreementsaccount_alias=C-0B7D2C30; cz_renewal_date=2026-09-10; cb_renewal_date=2026-09-15; is_multi_year=True; trusted=2026-09-15 | account_alias=C-0BBE3E60; cz_renewal_date=2027-09-26; cb_renewal_date=2026-09-26; is_multi_year=True; trusted=2026-09-26 | account_alias=C-0BCDB8C2; cz_renewal_date=2027-09-18; cb_renewal_date=2026-09-18; is_multi_year=True; trusted=2026-09-18 | account_alias=C-0D2AB865; cz_renewal_date=2026-09-10; cb_renewal_date=2026-09-22; is_multi_year=True; trusted=2026-09-22 | account_alias=C-0F5D2323; cz_renewal_date=2026-09-10; cb_renewal_date=2026-09-29; is_multi_year=True; trusted=2026-09-29
snapshot_date2026-09-05
total_arr_renewing_90d1048715.00
trust_rulemulti-year contracts: Chargebee is authoritative (ChurnZero known wrong); otherwise systems agree or Chargebee wins
window2026-09-05 to 2026-12-04
graded checkstotal_renewing · arr_at_risk · disagreements_flagged · trust_rule

Model's actual answer

show the model's full answer
90-Day Renewal Risk Brief

Missing data note: No as-of/start date was provided. All trusted renewal dates below fall between 2026-09-15 and 2026-11-24 — a 70-day span — so all 20 provided accounts are included as the 90-day cohort. No company names were provided; aliases are cited exactly as given.

Method
- Seat utilization = seats_used ÷ seats × 100
- 3-month usage trend = (2026-08 active_users − 2026-06 active_users) ÷ 2026-06 active_users × 100
- Date trust rule: per instruction, multi-year contracts are known to be wrong in ChurnZero, so Chargebee is trusted for is_multi_year=true accounts. Where both systems agree, date is used as corroborated.
- Risk rule (applied uniformly from the two provided signals only): High = utilization <30% OR 3-month trend ≤ −10%; Low = trend ≥0% AND utilization ≥70%; all others = Medium.

Date disagreements flagged: 5 of 20 accounts — C-0B7D2C30, C-0BCDB8C2, C-0D2AB865, C-0BBE3E60, C-0F5D2323. All are multi-year (24 or 36 months), so Chargebee dates are used for each. The other 15 accounts show identical dates in both systems; no disagreement.

Renewals

1. C-0B7D2C30 — Dana Mercer — $65,901 — Date used: 2026-09-15 [DISAGREEMENT: ChurnZero 2026-09-10 vs Chargebee 2026-09-15; Chargebee trusted, 36-mo multi-year] — Utilization: 274/476 = 57.6% — Trend: 97 → 84, −13 (−13.4%) — HIGH. Evidence: Active users fell 13.4% in 3 months.
2. C-0BCDB8C2 — Cole Ingram — $54,427 — Date used: 2026-09-18 [DISAGREEMENT: ChurnZero 2027-09-18 vs Chargebee 2026-09-18; Chargebee trusted, 36-mo multi-year] — Utilization: 232/424 = 54.7% — Trend: 127 → 110, −17 (−13.4%) — HIGH. Evidence: Active users fell 13.4% in 3 months.
3. C-0D2AB865 — Elena Sinclair — $38,022 — Date used: 2026-09-22 [DISAGREEMENT: ChurnZero 2026-09-10 vs Chargebee 2026-09-22; Chargebee trusted, 24-mo multi-year] — Utilization: 250/407 = 61.4% — Trend: 125 → 109, −16 (−12.8%) — HIGH. Evidence: Active users fell 12.8% in 3 months.
4. C-0BBE3E60 — Dana Mercer — $30,993 — Date used: 2026-09-26 [DISAGREEMENT: ChurnZero 2027-09-26 vs Chargebee 2026-09-26; Chargebee trusted, 24-mo multi-year] — Utilization: 74/114 = 64.9% — Trend: 39 → 33, −6 (−15.4%) — HIGH. Evidence: Active users fell 15.4% in 3 months, the steepest decline in the cohort.
5. C-0F5D2323 — Cole Ingram — $90,647 — Date used: 2026-09-29 [DISAGREEMENT: ChurnZero 2026-09-10 vs Chargebee 2026-09-29; Chargebee trusted, 24-mo multi-year] — Utilization: 111/390 = 28.5% — Trend: 20 → 18, −2 (−10.0%) — HIGH. Evidence: Only 28.5% of seats are used and active users fell 10.0%.
6. C-0EC6999D — Elena Sinclair — $79,419 — Date used: 2026-10-03 [both systems agree] — Utilization: 31/112 = 27.7% — Trend: 17 → 15, −2 (−11.8%) — HIGH. Evidence: Only 27.7% of seats are used and active users fell 11.8%.
7. C-0B20DB64 — Dana Mercer — $21,770 — Date used: 2026-10-07 [both systems agree] — Utilization: 214/378 = 56.6% — Trend: 294 → 294, 0 (0.0%) — MEDIUM. Evidence: Usage is flat but seat utilization is only 56.6%.
8. C-0BBC4E7A — Cole Ingram — $56,374 — Date used: 2026-10-10 [both systems agree] — Utilization: 228/337 = 67.7% — Trend: 142 → 139, −3 (−2.1%) — MEDIUM. Evidence: Active users slipped 2.1% with utilization below 70%.
9. C-0FD551AB — Elena Sinclair — $48,815 — Date used: 2026-10-14 [both systems agree] — Utilization: 210/376 = 55.9% — Trend: 123 → 126, +3 (+2.4%) — MEDIUM. Evidence: Usage rose 2.4% but seat utilization is only 55.9%.
10. C-0F9F8F13 — Dana Mercer — $46,230 — Date used: 2026-10-18 [both systems agree] — Utilization: 199/352 = 56.5% — Trend: 185 → 182, −3 (−1.6%) — MEDIUM. Evidence: Active users slipped 1.6% with utilization of 56.5%.
11. C-0BC34584 — Cole Ingram — $16,740 — Date used: 2026-10-22 [both systems agree] — Utilization: 327/494 = 66.2% — Trend: 104 → 106, +2 (+1.9%) — MEDIUM. Evidence: Usage rose 1.9% but seat utilization is 66.2%, below the 70% Low threshold.
12. C-0B7A7546 — Elena Sinclair — $35,062 — Date used: 2026-10-25 [both systems agree] — Utilization: 182/205 = 88.8% — Trend: 64 → 63, −1 (−1.6%) — MEDIUM. Evidence: Utilization is high at 88.8% but active users slipped 1.6%.
13. C-0B369871 — Dana Mercer — $85,128 — Date used: 2026-10-29 [both systems agree] — Utilization: 317/422 = 75.1% — Trend: 326 → 333, +7 (+2.1%) — LOW. Evidence: Utilization is 75.1% and active users rose 2.1%.
14. C-0B144C78 — Cole Ingram — $30,899 — Date used: 2026-11-02 [both systems agree] — Utilization: 169/224 = 75.4% — Trend: 101 → 106, +5 (+5.0%) — LOW. Evidence: Utilization is 75.4% and active users rose 5.0%.
15. C-0FC4DBB8 — Elena Sinclair — $94,732 — Date used: 2026-11-05 [both systems agree] — Utilization: 356/464 = 76.7% — Trend: 189 → 193, +4 (+2.1%) — LOW. Evidence: Utilization is 76.7% and active users rose 2.1%.
16. C-0D5BBE3A — Dana Mercer — $39,740 — Date used: 2026-11-09 [both systems agree] — Utilization: 85/102 = 83.3% — Trend: 88 → 91, +3 (+3.4%) — LOW. Evidence: Utilization is 83.3% and active users rose 3.4%.
17. C-0FB9D5AF — Cole Ingram — $63,158 — Date used: 2026-11-13 [both systems agree] — Utilization: 144/199 = 72.4% — Trend: 173 → 176, +3 (+1.7%) — LOW. Evidence: Utilization is 72.4% and active users rose 1.7%.
18. C-0B344485 — Elena Sinclair — $64,384 — Date used: 2026-11-16 [both systems agree] — Utilization: 224/287 = 78.0% — Trend: 238 → 244, +6 (+2.5%) — LOW. Evidence: Utilization is 78.0% and active users rose 2.5%.
19. C-0CB2C1B4 — Dana Mercer — $40,628 — Date used: 2026-11-20 [both systems agree] — Utilization: 386/473 = 81.6% — Trend: 47 → 49, +2 (+4.3%) — LOW. Evidence: Utilization is 81.6% and active users rose 4.3%.
20. C-22170CA1 — Cole Ingram — $45,646 — Date used: 2026-11-24 [both systems agree] — Utilization: 251/294 = 85.4% — Trend: 143 → 146, +3 (+2.1%) — LOW. Evidence: Utilization is 85.4% and active users rose 2.1%.

Totals (arithmetic shown)

- High-risk ARR: $65,901 + $54,427 + $38,022 + $30,993 + $90,647 + $79,419 = $359,409
- Medium-risk ARR: $21,770 + $56,374 + $48,815 + $46,230 + $16,740 + $35,062 = $224,991
- Low-risk ARR: $85,128 + $30,899 + $94,732 + $39,740 + $63,158 + $64,384 + $40,628 + $45,646 = $464,315
- Total ARR renewing: $359,409 + $224,991 + $464,315 = $1,048,715
- ARR at risk (High-risk only, per the rule above): $359,409, which is $359,409 ÷ $1,048,715 = 34.3% of renewing ARR. If Medium is also counted as exposed, at-risk ARR is $359,409 + $224,991 = $584,400.
ticket-theme-synthesisPASS 1.00
Plain-English question: What are the real themes in this quarter's support tickets, ignoring the existing tags?
Prompt — what the model saw
Synthesize the quarter's support tickets into 4 to 7 themes. Do not trust existing tags; read the text. Per theme: name, count, share, distinct accounts, ARR affected, two ticket ids, one-line recommendation. Separate broad patterns from single-account noise and rank by ARR exposure, not volume.
cost $0.2201 · 29s · in 22,258 / out 5,934 tok
✓ theme_count_range✓ counts_verifiable✓ single_account_flagged✓ arr_ranked✓ ticket_ids

✓ Correct Answer (ground truth)

Plain-English answer: 80 tickets in 4-7 themes — points-not-posting leads (20 tickets, 25%, $31,100 ARR affected) — each with counts, distinct accounts, ARR, and a recommendation.
The answer in one line: total_tickets = 80; quarter_window = 2026-06-01 through 2026-08-31
total_tickets80
noteexisting tags are noisy/unreliable; classify by text
quarter_window2026-06-01 through 2026-08-31
ranking_by_arr_exposurehris_provisioning, redemption_failures, billing_dispute, points_not_posting, slack_sync
sample_ticket_ids80 items — e.g. IC-460004, IC-460016, IC-460001, IC-460041
single_account_noise_themebilling_dispute
themes{"points_not_posting": {"count": 20, "share_pct": 25.0, "distinct_accounts": 9, "arr_affected": "31100.00", "example_ticket_ids": ["IC-460001", "IC-460002"], "single_account": false}, "slack_sync": {"
graded checkstheme_count_range · counts_verifiable · single_account_flagged · arr_ranked · ticket_ids

Model's actual answer

show the model's full answer
Basis: 80 tickets. Themes read from body_text, not tag. Ranked by ARR affected = ARR summed once per distinct account in the theme (not per ticket). Shares = count / 80.

**Broad patterns (multi-account), ranked by ARR exposure**

1. **HRIS provisioning not creating new-hire accounts**
- Count / share: 12 / 80 = 15.0%
- Distinct accounts: 3 — C-0B2213A9, C-0F6C0F34, C-0DDFC9A7
- ARR affected: $36,000 + $30,000 + $48,000 = $114,000
- Tickets: IC-460059, IC-460055
- Recommendation: Fix silent HRIS sync skips and add provisioning-log alerting when new hires are not created.

2. **Redemption / gift-card checkout failures**
- Count / share: 18 / 80 = 22.5%
- Distinct accounts: 7 — C-0CEF69FD, C-0B827671, C-0FCCD2DF, C-0F876796, C-14264ABD, C-0D9CA315, C-0B0F1BAB
- ARR affected: $8,900 + $10,700 + $9,600 + $8,700 + $11,000 + $9,600 + $10,300 = $68,800
- Tickets: IC-460025, IC-460030
- Recommendation: Fix checkout failure and ensure failed redemptions do not deduct points and do deliver the gift-card code/email.

3. **Recognition points not posting to balances**
- Count / share: 20 / 80 = 25.0%
- Distinct accounts: 9 — C-0D3278C7, C-0BF20542, C-0D0B047C, C-0BE96399, C-0D284E42, C-0D6CC8E3, C-21FEBCBB, C-0DD0626C, C-0B2895EF
- ARR affected: $3,500 + $4,500 + $4,500 + $2,700 + $3,400 + $4,200 + $2,900 + $2,500 + $2,900 = $31,100
- Tickets: IC-460004, IC-460016
- Recommendation: Trace delivered recognitions to balance posting and backfill missing points; highest volume theme despite lower ARR.

4. **Slack integration failures (sync, auth, slash command)**
- Count / share: 14 / 80 = 17.5%
- Distinct accounts: 4 — C-0B843542, C-10A56B0F, C-0BA71F12, C-8C2E8F00
- ARR affected: $4,400 + $5,400 + $3,900 + $5,200 = $18,900
- Tickets: IC-460041, IC-460047
- Recommendation: Fix Slack sync-toggle reset, re-auth persistence, and slash-command errors as one integration-reliability workstream.

**Single-account noise (separate from broad patterns)**

5. **Billing: seat-count / tier-price invoice errors**
- Count / share: 16 / 80 = 20.0%
- Distinct accounts: 1 — C-0E9C27D1 only
- ARR affected: $52,000 (single account, counted once; not $52,000 × 16)
- Tickets: IC-460071, IC-460069
- Recommendation: Resolve C-0E9C27D1's repeated seat-count (150 vs 200) and wrong-tier renewal billing directly with that account; do not treat as a platform-wide billing pattern on this data alone.

Check: 12 + 18 + 20 + 14 + 16 = 80 tickets; 15.0% + 22.5% + 25.0% + 17.5% + 20.0% = 100%.
marketing 5 tests
case-study-matchPASS 1.00
Plain-English question: Which three existing customers are the best case-study matches for this prospect?
Prompt — what the model saw
Given the prospect, rank the three most similar existing customers with public case studies for use as social proof. Similarity on industry, size band, use case, region. Cite the fields that drove each match. Do not name any customer without a case study and never include billing data or contact names.
cost $0.0206 · 22s · in 670 / out 688 tok
✓ top3✓ decoys_excluded✓ cites_fields✓ no_billing

✓ Correct Answer (ground truth)

Plain-English answer: Three ranked matches on industry, size, use case, and region (e.g. C-11C31562), with the higher-scoring decoy explicitly excluded for a mismatched field.
The answer in one line: prospect_alias = C-82AF3719
constraintsno_billing_data, no_contact_names, only_customers_with_has_case_study_true_named
decoys_excludedcompany_alias=C-D6217CAA; similarity_score=4; reason=no_case_study | company_alias=C-C153868F; similarity_score=4; reason=no_case_study
prospect_aliasC-82AF3719
prospect_fields_usedindustry, size_band, use_case, region
top3company_alias=C-11C31562; similarity_score=3 | company_alias=C-64171065; similarity_score=3 | company_alias=C-A13C193D; similarity_score=3
graded checkstop3 · decoys_excluded · cites_fields · no_billing

Model's actual answer

show the model's full answer
Prospect: C-82AF3719 — Technology, Mid-Market, employee_recognition, NA-West

Scored only customers with `has_case_study=true`. Similarity = matching fields / 4 fields (industry, size_band, use_case, region).

Top 3 — all tied at 3/4:

1. C-64171065 — 3/4 = 75%
   - Match: industry=Technology, size_band=Mid-Market, use_case=employee_recognition
   - Mismatch: region=NA-East (prospect: NA-West)

1. C-11C31562 — 3/4 = 75%
   - Match: size_band=Mid-Market, use_case=employee_recognition, region=NA-West
   - Mismatch: industry=Manufacturing (prospect: Technology)

1. C-A13C193D — 3/4 = 75%
   - Match: industry=Technology, size_band=Mid-Market, region=NA-West
   - Mismatch: use_case=retention (prospect: employee_recognition)

No case-study customer matches all 4 fields.
channel-attribution-mathPASS 1.00
Plain-English question: Compute cost per SQM, cost per SQO, and pipeline per dollar for every channel over six months.
Prompt — what the model saw
Compute channel performance for the trailing 6 months. Paid: spend, SQMs, SQOs, cost per SQM, cost per SQO, SQM-to-SQO rate, pipeline amount, pipeline per dollar. Organic: volume, SQO rate, pipeline. Report a channel with spend and zero SQMs as undefined, not zero. Flag rows where SQO date precedes SQM date. Finish with a reallocation recommendation and your confidence given sample sizes.
cost $0.1099 · 41s · in 3,641 / out 4,327 tok
✓ undefined_channel✓ undefined_not_zero✓ flagged_rows✓ best_channel✓ reallocation

✓ Correct Answer (ground truth)

Plain-English answer: LinkedIn Ads at $960/SQM and $3,000/SQO, organic search at $90,000 pipeline on zero spend, paid social flagged at $18,000 with zero SQMs, and a channel winner named.
The answer in one line: best_pipeline_per_dollar_channel = paid_search
best_pipeline_per_dollar_channelpaid_search
confidence_notewebinars and referral have small SQM samples (<20); treat rates as directional.
flagged_rows_sqo_before_sqmCT-000041, CT-000044
months2026-03, 2026-04, 2026-05, 2026-06, 2026-07, 2026-08
per_channel{"linkedin_ads": {"spend_usd": 24000, "sqms": 25, "sqos": 8, "pipeline_amount": 96000, "cost_per_sqm": 960.0, "cost_per_sqo": 3000.0, "sqm_to_sqo_rate": 0.32, "pipeline_per_dollar": 4.0}, "organic_sea
reallocation_recommendationfrom=paid_social; to=paid_search; rationale=paid_social has spend but zero SQMs (cost per SQM undefined, not zero); paid_search has the highest pipeline per dollar.
undefined_channelspaid_social
graded checksundefined_channel · undefined_not_zero · flagged_rows · best_channel · reallocation

Model's actual answer

show the model's full answer
Trailing 6 months = 2026-03 through 2026-08, inclusive. All 122 contacts in contacts_first_touch.csv have sqm_date in that window.

Spend arithmetic:
- paid_search: 6 × $6,000 = $36,000
- linkedin_ads: 6 × $4,000 = $24,000
- paid_social: 6 × $3,000 = $18,000
- webinars: 6 × $1,500 = $9,000

Paid channels — defined as channels present in channel_spend.csv

paid_search
- SQMs: 40
- SQOs: 18 (SQO = sqo_date populated)
- Cost per SQM: $36,000 / 40 = $900.00
- Cost per SQO: $36,000 / 18 = $2,000.00
- SQM-to-SQO rate: 18 / 40 = 45.0%
- Pipeline: 18 × $40,000 = $720,000
- Pipeline per dollar: $720,000 / $36,000 = $20.00

linkedin_ads
- SQMs: 25
- SQOs: 8
- Cost per SQM: $24,000 / 25 = $960.00
- Cost per SQO: $24,000 / 8 = $3,000.00
- SQM-to-SQO rate: 8 / 25 = 32.0%
- Pipeline: 8 × $12,000 = $96,000
- Pipeline per dollar: $96,000 / $24,000 = $4.00

paid_social
- SQMs: 0
- SQOs: 0
- Cost per SQM: $18,000 / 0 = undefined (spend with zero SQMs — not zero, per instruction)
- Cost per SQO: $18,000 / 0 = undefined
- SQM-to-SQO rate: 0 / 0 = undefined
- Pipeline: $0
- Pipeline per dollar: $0 / $18,000 = $0.00

webinars
- SQMs: 12
- SQOs: 5
- Cost per SQM: $9,000 / 12 = $750.00
- Cost per SQO: $9,000 / 5 = $1,800.00
- SQM-to-SQO rate: 5 / 12 = 41.7%
- Pipeline: 5 × $12,000 = $60,000
- Pipeline per dollar: $60,000 / $9,000 = $6.67

Organic channels — defined as channels in contacts_first_touch.csv with no spend in channel_spend.csv. No spend data was provided for these, so no cost metrics are computed.

organic_search
- Volume (SQMs): 30
- SQOs: 10
- SQO rate: 10 / 30 = 33.3%
- Pipeline: 10 × $9,000 = $90,000

referral
- Volume (SQMs): 15
- SQOs: 6
- SQO rate: 6 / 15 = 40.0%
- Pipeline: 6 × $8,000 = $48,000

Data-quality flags — SQO date precedes SQM date
- CT-000044, linkedin_ads: sqm_date 2026-07-23, sqo_date 2026-07-18 (SQO 5 days before SQM), pipeline_amount $12,000
- CT-000041, linkedin_ads: sqm_date 2026-06-14, sqo_date 2026-06-09 (SQO 5 days before SQM), pipeline_amount $12,000

Both are counted as SQOs above, as given. Sensitivity if excluded: linkedin_ads SQOs = 6, cost per SQO = $24,000 / 6 = $4,000.00, SQM-to-SQO = 6 / 25 = 24.0%, pipeline = $72,000, pipeline per dollar = $3.00.

Reallocation recommendation
Move paid_social spend first. $18,000 over 6 months produced 0 SQMs and $0 pipeline in the data provided; cost per SQM and cost per SQO are undefined, not merely high. Redirect that budget to paid_search, which has the largest sample (40 SQMs) and the best economics on every paid metric: $900.00 cost per SQM, $2,000.00 cost per SQO, 45.0% SQM-to-SQO, $20.00 pipeline per dollar — 3.0× webinars ($6.67) and 5.0× linkedin_ads ($4.00). Hold webinars funding flat pending more volume rather than scaling it: its $1,800.00 cost per SQO is the lowest paid figure, but it rests on 5 SQOs from 12 SQMs. Trim linkedin_ads last and only partially: it converts (8 SQOs), just at worse unit economics, and 2 of its 8 SQOs carry the date flags above.

Confidence: low to moderate. Sample sizes are small — paid_search n=40, linkedin_ads n=25, webinars n=12, paid_social n=0, organic_search n=30, referral n=15. At n=12, webinars' 41.7% rate is 5 events; one SQO either way moves it ~8 points, so its ranking versus paid_search (45.0% on n=40) is not reliably separated. The paid_search advantage is directionally consistent across cost per SQM, cost per SQO, conversion rate, and pipeline per dollar, which raises confidence in the direction, but no statistical test was requested or run, and pipeline_amount is a fixed per-SQO value within each channel ($40,000 / $12,000 / $12,000 / $9,000 / $8,000), so pipeline per dollar is driven by SQO count, not by observed deal-size variation. Missing data: no closed-won, revenue, or SQO-to-close data was provided, so this is a pipeline-efficiency recommendation only, not a revenue recommendation.
competitor-battlecard-updatePASS 1.00
Plain-English question: Update the Rivally battlecard with sourced pricing and where each side wins.
Prompt — what the model saw
Update the battlecard for the competitor. Sections: one-line positioning, pricing with source and date (newer source wins, note the conflict), where they win, where we win, objections and responses, recent changes, our 12-month win/loss record against them. Cite a snippet id for every factual claim. Rep opinion on a call is not a fact about the competitor. Anything from the old card you cannot re-source gets marked unverified.
cost $0.0891 · 41s · in 3,797 / out 3,285 tok
✓ competitor_named✓ pricing✓ win_loss✓ unverified_marked✓ citations

✓ Correct Answer (ground truth)

Plain-English answer: Rivally at $7 per user per month (the August source wins over the stale $5, conflict noted), a 13-7 win/loss record, and two unverified old claims flagged.
The answer in one line: competitor = Rivally
citation_ids_available25 items — e.g. S01, S02, S03, S04
competitorRivally
pricingvalue_per_user_per_month=7; source_snippet=S17; source_date=2026-08-12
rep_opinion_snippet_ids_not_factsS09, S21
unverified_old_claimsRivally lacks a Slack integration., Rivally was acquired by WorkHuman in 2025.
win_loss_12mowins=13; losses=7
graded checkscompetitor_named · pricing · win_loss · unverified_marked · citations

Model's actual answer

show the model's full answer
# Updated Battlecard: Rivally

## One-Line Positioning
Points-based recognition [S02], with an engaging recognition feed [S16].

Old-card addition "for mid-market" is **unverified** — no snippet in the provided data sources that positioning claim. A mid-market reviewer is mentioned [S04], but that does not source the positioning.

## Pricing
Current list (newer source wins): Recognition Starter is **$7 per user/month, annual billing required** — Rivally pricing page, 2026-08-12 [S17].

Conflict noted:
- $5 per user/month, annual billing required — pricing page, 2026-01-20 [S03]
- Pricing page still showed $5 per user/month for Recognition Starter — 2026-04-01 [S08]
- Updated to $7 per user/month — 2026-08-12 [S17]. This supersedes [S03] and [S08].

Deal-specific quotes in the data (not list price):
- $6.50/user/mo quoted to a 500-seat prospect, annual term — 2026-06-02 [S13]
- $7/user/mo list, with 15% discount offered for a 3-year term — 2026-08-14 [S18]. Arithmetic: $7.00 × (1 − 0.15) = $7.00 × 0.85 = **$5.95/user/mo** at that discount [S18].

## Where They Win
- EU / distributed teams: "strong for distributed EU teams, multi-language support praised" — EU enterprise reviewer [S12]; EU data residency generally available [S15]; pitched EU data residency to a prospect [S05].
- Fast setup and Slack: "setup took under a week, Slack integration worked out of the box" — mid-market reviewer [S04]. This contradicts the old card (see Unverified / Corrected below).
- Recognition feed: praised as a points-based recognition feed [S02]; "recognition feed is engaging" [S16].
- Support: "support response time praised (under 4 hours)" [S22].

## Where We Win
Stated only where the provided data supports it:
- Analytics depth: an 800-seat prospect picked Bonusly over Rivally citing analytics depth [S25]. Supporting Rivally gaps in the data: "limited analytics" [S02]; "reporting dashboards are basic compared to enterprise tools" [S07]; "analytics exports are CSV-only," which made migration off Rivally hard [S20].
- Admin tooling: "admin tooling lags peers" [S16]; "admin console still lacks bulk recognition editing" [S24].
- Enterprise provisioning: "Rivally lacks SCIM provisioning; manual user management is painful" — enterprise reviewer [S10].
- EMEA rewards breadth: "rewards catalog in EMEA is thinner than US catalog" [S14].

No other Bonusly advantages are claimed — no other comparative data was provided.

## Objections and Responses
- Objection: "Rivally is $5/user/month." Response: That was the pricing-page price on 2026-01-20 [S03] and still shown 2026-04-01 [S08]; the pricing page updated Recognition Starter to $7/user/month, annual billing required, on 2026-08-12 [S17].
- Objection: "Rivally is stronger for our EU teams." Response: Acknowledge the sourced strengths — multi-language support praised [S12] and EU data residency generally available [S15] — then probe the sourced gaps: EMEA rewards catalog thinner than the US catalog [S14] and basic reporting dashboards [S07].
- Objection: "Rivally will be faster to roll out." Response: Setup under a week with Slack working out of the box is sourced [S04]; ask how user provisioning will be handled, since an enterprise reviewer reports no SCIM provisioning and painful manual user management [S10].
- Objection: "Rivally's reporting is enough for us." Response: Test against the sourced limits: limited analytics [S02], basic reporting dashboards [S07], CSV-only analytics exports [S20] — and note an 800-seat prospect chose Bonusly over Rivally citing analytics depth [S25].

## Recent Changes
- Series C of $40M led by Northgate Ventures announced — 2025-11-04 [S01].
- 'Rivally Pulse', a lightweight engagement survey add-on, launched — 2026-03-05 [S06]; Pulse later exited beta, "priced as add-on, not bundled" — 2026-09-01 [S23].
- Hired ex-Workday VP EMEA to lead European expansion — 2026-05-09 [S11].
- Opened Dublin office; EU data residency generally available — 2026-07-01 [S15].
- Recognition Starter price raised to $7/user/month — 2026-08-12 [S17].
- Microsoft Teams app v2 announced in public preview — 2026-08-19 [S19].

## Our 12-Month Win/Loss Record Against Rivally
Window in the provided deal data: 2025-09 through 2026-08 (12 months). All 20 listed deals name Rivally as competitor.

Arithmetic: 13 wins + 7 losses = 20 deals. Win rate = 13 ÷ 20 = 0.65 = **65%**. Loss rate = 7 ÷ 20 = 0.35 = 35%.

Wins (13): Deal-A9FD43, Deal-7AA785, Deal-44C524, Deal-0D0CD6, Deal-D5B790, Deal-5C636E, Deal-67BE14, Deal-1B6969, Deal-F03E7B, Deal-072E31, Deal-F65C8F, Deal-E46EAB, Deal-1D2392

Losses (7): Deal-7767F5, Deal-5645A5, Deal-C6FFAA, Deal-D263E0, Deal-935746, Deal-9066A6, Deal-72A02F

## Unverified / Corrected From the Old Card
- "Rivally was acquired by WorkHuman in 2025" — **unverified**. No snippet in the provided data sources this. Removed.
- "Rivally lacks a Slack integration" — **contradicted**, not merely unverified: a reviewer reports the "Slack integration worked out of the box" [S04]. Corrected above.
- "Pricing: starts at $5 per user/month, annual billing (as of 2026-01)" — superseded by the 2026-08-12 pricing page at $7 [S17]; conflict retained in Pricing above.
- "Strong in EU enterprise with multi-language support" — re-sourced: [S12]. Retained.
- "Points-based recognition for mid-market" — "points-based recognition" re-sourced [S02]; "for mid-market" **unverified** as a positioning claim.

## Explicitly Excluded (Rep Opinion, Not Competitor Fact)
- AE opinion that Rivally's UI is clunky — the snippet itself states it is "Not verified against the product" [S09]. Not used.
- AE opinion that Rivally is discounting aggressively — the snippet itself states "Rep opinion, not confirmed" [S21]. Not used. The only discount claim used is the deal mention of a 15% discount offered for a 3-year term [S18].

Missing data: no snippets were provided on Rivally's security certifications, contract terms beyond those cited, or any Bonusly pricing — none are claimed here.
nurture-sequence-diagnosisPASS 1.00
Plain-English question: What is broken in each email sequence — rates, overlaps, tracking errors?
Prompt — what the model saw
Review the sequences. Per sequence: sent, open rate, reply rate, meeting rate, weakest step. Identify tracking errors (opened above sent), audience overlap between sequences, and the failure mode for anything under 2 percent reply. One change per weak sequence and which to fix first. Under 350 words.
cost $0.1537 · 36s · in 24,386 / out 2,397 tok
✓ tracking_error✓ overlap_pair✓ weakest_step✓ fix_priority

✓ Correct Answer (ground truth)

Plain-English answer: Per-sequence open, reply, and meeting rates with the weakest step (all three weakest at step 3), an opened-above-sent tracking error, and audience overlap flagged.
The answer in one line: fix_priority_first = Cold Outbound - HR Leaders
all_pairwise_overlaps['Cold Outbound - HR Leaders', 'Cold Outbound - People Ops'], ['Expansion Nurture', 'New Logo Nurture']
audience_overlap_pairCold Outbound - HR Leaders, Cold Outbound - People Ops
failure_modeCold Outbound - HR Leaders=opens healthy (~37%) but replies near zero from step 1 -> body/CTA not landing, not a deliverability or subject-line problem
fix_priority_firstCold Outbound - HR Leaders
per_sequence{"New Logo Nurture": {"sent": 1386, "opened": 490, "replied": 90, "meetings": 27, "open_rate": 0.3535, "reply_rate": 0.0649, "meeting_rate": 0.0195, "weakest_step": 3}, "Expansion Nurture": {"sent": 8
tracking_errorsequence=Expansion Nurture; step=2; sent=300; opened=340; issue=opened exceeds sent
under_2pct_replyCold Outbound - HR Leaders
graded checkstracking_error · overlap_pair · weakest_step · fix_priority

Model's actual answer

show the model's full answer
Rates = total / total sent (sum of steps).

Per sequence

New Logo Nurture: sent 1,386 (500+458+428); open 490/1,386 = 35.4%; reply 90/1,386 = 6.5%; meeting 27/1,386 = 1.9%. Weakest step: 3 — reply 18/428 = 4.2%, meeting 6/428 = 1.4% (lowest).

Expansion Nurture: sent 875 (300+300+275); open 565/875 = 64.6% (invalid, see below); reply 59/875 = 6.7%; meeting 12/875 = 1.4%. Weakest step: 3 — reply 12/275 = 4.4%, meeting 3/275 = 1.1%.

Cold Outbound - HR Leaders: sent 1,785 (600+595+590); open 545/1,785 = 30.5%; reply 8/1,785 = 0.45%; meeting 0/1,785 = 0%. Weakest step: 3 — reply 1/590 = 0.17%.

Cold Outbound - People Ops: sent 1,163 (400+386+377); open 340/1,163 = 29.2%; reply 29/1,163 = 2.5%; meeting 6/1,163 = 0.5%. Weakest step: 3 — reply 6/377 = 1.6%, meeting 1/377 = 0.3%.

Tracking error
Expansion Nurture step 2: opened 340 > sent 300 = 113.3%. Impossible; its sequence open rate (64.6%) is unreliable. No other step has opened > sent.

Audience overlap
Found in audiences.csv (examples, exact keys):
- CT-000301 and CT-000624: both New Logo Nurture and Expansion Nurture.
- CT-000849, CT-001255, CT-001105, CT-001103, CT-001153: both Cold Outbound - HR Leaders and Cold Outbound - People Ops.
So the same contacts are receiving competing cold sequences, and nurture audiences are not mutually exclusive. Unique-contact counts per sequence are missing from sequences.csv, so overlap % cannot be computed from sent totals.

Failure mode (<2% reply)
Only Cold Outbound - HR Leaders qualifies at sequence level: 0.45%. Opens are 30.5% but replies are 8/1,785 and meetings 0 — this is a message/offer/CTA failure after the open, not an open/deliverability failure. Cold Outbound - People Ops step 3 (1.6%) shows the same late-step decay.

One change per weak sequence
- Cold Outbound - HR Leaders: rewrite step 1 offer/CTA only (keep subject/open mechanics); opens are adequate, replies are not.
- Cold Outbound - People Ops: replace step 3 with a new final angle rather than a repeat bump; reply fell 3.5% -> 2.3% -> 1.6%.

Fix first: Cold Outbound - HR Leaders — largest volume (1,785 sent), 0.45% reply, 0 meetings.
weekly-marketing-goals-rollupPASS 1.00
Plain-English question: Are we on pace on the quarter's marketing goals?
Prompt — what the model saw
Produce the weekly marketing goals update for the active quarter: SQMs, SQOs, DS2s, closed-lost MIA rate, same-quarter close count, active pipeline coverage against target. For each: QTD actual, target, delta, pace (on/behind/ahead given days elapsed). One paragraph on what moved this week.
cost $0.0433 · 29s · in 1,572 / out 1,441 tok
✓ metrics_present✓ deltas✓ coverage_ratio✓ pace✓ paragraph

✓ Correct Answer (ground truth)

Plain-English answer: SQMs ahead of pace, SQOs on pace, DS2s behind (35 short of expected at day 66 of 92), with deltas and same-quarter closes.
The answer in one line: days_in_quarter = 92; days_elapsed = 66; coverage_ratio_active_pipeline = 0.75; elapsed_fraction = 0.7174 (+1 more below)
days_in_quarter92
coverage_ratio_active_pipeline0.75
days_elapsed66
elapsed_fraction0.7174
metrics{"SQMs": {"qtd_actual": 230, "target": 300, "delta": -70, "expected_at_day_elapsed": 215.22, "pace": "ahead"}, "SQOs": {"qtd_actual": 84, "target": 120, "delta": -36, "expected_at_day_elapsed": 86.09,
pace_ruleexpected = target * days_elapsed/days_in_quarter; within +/-5% of expected = on, above = ahead, below = behind; lower_better metrics invert.
this_week_paragraph_required1
graded checksmetrics_present · deltas · coverage_ratio · pace · paragraph

Model's actual answer

show the model's full answer
Weekly Marketing Goals Update — Q3-2026

Quarter pace basis: 66 days elapsed / 92 days in quarter = 71.74% elapsed. Pro-rated expected = target × 66 / 92. Pace is ahead if QTD actual exceeds pro-rated expected, behind if below it. The closed-lost MIA rate is a rate, not a cumulative count, so it is compared directly to target.

| Metric | QTD actual | Target | Delta | Pace |
|---|---|---|---|---|
| SQMs | 230 | 300 | 230 − 300 = −70 (76.67% of target) | Ahead — pro-rated expected = 300 × 71.74% = 215.22; 230 − 215.22 = +14.78 |
| SQOs | 84 | 120 | 84 − 120 = −36 (70.00% of target) | Behind — pro-rated expected = 120 × 71.74% = 86.09; 84 − 86.09 = −2.09 |
| DS2s | 40 | 75 | 40 − 75 = −35 (53.33% of target) | Behind — pro-rated expected = 75 × 71.74% = 53.80; 40 − 53.80 = −13.80 |
| closed_lost_mia_rate | 20.00% (5 / 25) | 10.00% | 20.00% − 10.00% = +10.00pp | Behind (lower is better) — rate is 2.0× target |
| same_quarter_closes | 10 | 20 | 10 − 20 = −10 (50.00% of target) | Behind — pro-rated expected = 20 × 71.74% = 14.35; 10 − 14.35 = −4.35 |
| active_pipeline | 3,000,000 | 4,000,000 | 3,000,000 − 4,000,000 = −1,000,000 (75.00% coverage) | Ahead on pace, below full target — pro-rated expected = 4,000,000 × 71.74% = 2,869,565.22; 3,000,000 − 2,869,565.22 = +130,434.78 |

What moved this week

Data is missing to determine what moved this week: the files provided contain only QTD actuals, targets, and quarter dates (66 of 92 days elapsed), with no prior-week snapshot or week-over-week values. On the current QTD snapshot alone, SQMs and active_pipeline are above pro-rated pace, while SQOs, DS2s, same_quarter_closes, and the closed_lost_mia_rate (5 of 25 closed-lost) are behind pace or target; no weekly change can be stated from this data.
communication 5 tests
ceo-slack-compressionPASS 1.00
Plain-English question: Compress the forecast analysis into one Slack message the CEO can read in ten seconds.
Prompt — what the model saw
Compress the forecast analysis below into a single Slack message for the CEO. Rules: at most three sentences; under 80 words; lead with the decision or the so-what; keep every distinct number that matters and the one caveat that changes the conclusion; cut any sentence that does not carry a claim, a number, or a directive. No em dashes, no throat-clearing phrases.
cost $0.0243 · 13s · in 558 / out 899 tok
✓ max_3_sentences✓ under_80_words✓ weighted_present✓ caveat_present✓ no_em_dash

✓ Correct Answer (ground truth)

Plain-English answer: Three sentences, under 80 words, decision first, every distinct number kept plus the one caveat — all checked by deterministic text rules.
The answer in one line: weighted_number = 115976; must_keep = data-quality caveat — deleting the bad news fails
weighted_number115,976
limitmax 3 sentences, max 80 words, no em-dashes
must_keepdata-quality caveat — deleting the bad news fails
graded checksmax_3_sentences · under_80_words · weighted_present · caveat_present · no_em_dash

Model's actual answer

show the model's full answer
Q3 forecast is $115,976.75, from $44,729 COMMIT (7 deals) plus $71,247.75 (35% of $203,565 BEST_CASE, 24 deals), with 23 PIPELINE deals at $0. Only 54 of 86 deals close in quarter; 32 deals totaling $227,575, including COMMIT Deal-D348E1 ($13,770 on 2026-10-15), are excluded. Do not treat this as reliable: why-buys is empty on all 7 in-quarter COMMIT deals, so the committed $44,729 has no documented reason to buy.
stalled-deal-followup-emailPASS 1.00
Plain-English question: Write a follow-up email for this stalled deal — under 120 words, one prior touch, one ask.
Prompt — what the model saw
Write a follow-up email for this stalled deal. Under 120 words. Reference one specific prior touch from the engagement history, make one clear ask, and match the tone of the prior thread. No invented details.
cost $0.0171 · 21s · in 387 / out 574 tok
✓ word_cap✓ references_touch✓ single_ask

✓ Correct Answer (ground truth)

Plain-English answer: An email that references the July 28 milestone-awards demo, makes one clear ask, and invents nothing — Deal-0D2F7A, 31 days since last engagement.
The answer in one line: amount = 5100; word_cap = 120; days_since_last_engagement = 31; deal_alias = Deal-0D2F7A (+4 more below)
amount5,100
days_since_last_engagement31
word_cap120
allowed_facts14 items — e.g. Deal-0D2F7A, $5,100, DS3, Alex Franklin
allowed_newENG-00001, ENG-00002, ENG-00003, ENG-00004, ENG-00005
deal_aliasDeal-0D2F7A
must_referenceengagement_id=ENG-00002; date=2026-07-28; type=meeting; why=most substantive prior touch; concrete prospect reactions to milestone awards and points catalog
ownerAlex Franklin
single_cta1
stageDS3
tonewarm, concise, professional (matches prior thread)
graded checksword_cap · references_touch · single_ask

Model's actual answer

show the model's full answer
Subject: Follow-up on the July 28 demo pricing

Hi team,

Following up on my August 5 email, which recapped the July 28 demo and included pricing for 150 seats. In that demo, the People team reacted strongly to automated milestone awards and the points catalog, and asked for pricing.

Could you let me know whether that pricing works to move forward?

Thanks,
Alex Franklin
funnel-digest-paragraphPASS 1.00
Plain-English question: Write the weekly GTM digest — one paragraph per team, every number traceable.
Prompt — what the model saw
Write the weekly GTM digest: one paragraph per team (Marketing, Sales, CS, PLG) in Bonusly brand voice. Every number must trace to the extracts. No customer contact names, no employee performance commentary, no compensation. Four paragraphs, under 400 words total.
cost $0.0144 · 21s · in 245 / out 463 tok
✓ four_paragraphs✓ word_cap✓ teams

✓ Correct Answer (ground truth)

Plain-English answer: Four paragraphs (Marketing, Sales, CS, PLG) using only the allowed numbers, no contact names, no performance commentary, under 400 words.
The answer in one line: word_cap = 400
word_cap400
allowed_numbers16 items — e.g. 2, 3, 9, 08
bannedcustomer contact names, employee performance commentary, compensation
required_paragraphsMarketing, Sales, CS, PLG
graded checksfour_paragraphs · word_cap · teams

Model's actual answer

show the model's full answer
Marketing came in at 46 SQMs against a target of 52, a shortfall of 6 (52 − 46 = 6). The webinar channel delivered 18 of those SQMs. Cost per SQM on paid search held at $150.

Sales converted 14 SQOs and set 9 DS2 meetings. New pipeline created was $310,000. Same-quarter close count for the week was 3.

CS saved 2 renewals this week. Team NPS moved to 61. There are 3 open red-flag accounts heading into next week.

PLG added 412 new signups with activation at 31%. 38 companies reached the aha moment of 10 recognition gives.
partner-digest-sectionPASS 1.00
Plain-English question: Write the partner digest so every partner in the roster appears exactly once.
Prompt — what the model saw
Write the partner digest for the period. Every partner in the roster appears once: active plays get status with numbers, quiet partners get one line. Partner-sourced pipeline count and amount must match the deal data. Nothing omitted, nothing added.
cost $0.0205 · 21s · in 433 / out 734 tok
✓ all_partners✓ pipeline_amount✓ quiet_one_line

✓ Correct Answer (ground truth)

Plain-English answer: Four active partners with status and numbers, four quiet partners with one line each, and partner-sourced pipeline that reconciles to the deal data.
The answer in one line: rule = every roster partner named exactly once; quiet partners get one line; no partners outside roster.
active_no_dealsWorkWell Group
partner_sourced_pipelinecount=5; amount=550000
per_partner_pipeline{"Apex Rewards Co": {"count": 2, "amount": 275000}, "HRCloud Partners": {"count": 1, "amount": 140000}, "CultureBridge": {"count": 2, "amount": 135000}}
quiet_partnersRecogniq, KudosWave, PeopleFirst Advisors, TotalPerk
rosterpartner_name=Apex Rewards Co; status=active | partner_name=HRCloud Partners; status=active | partner_name=CultureBridge; status=active | partner_name=WorkWell Group; status=active | partner_name=Recogniq; status=quiet | partner_name=KudosWave; status=quiet … (+2 more)
ruleevery roster partner named exactly once; quiet partners get one line; no partners outside roster.
graded checksall_partners · pipeline_amount · quiet_one_line

Model's actual answer

show the model's full answer
Partner Digest — 2026-08-24 to 2026-09-04

Partner-sourced pipeline total: 5 deals, $550,000
Arithmetic: $180,000 + $95,000 + $140,000 + $60,000 + $75,000 = $550,000

Active partners

Apex Rewards Co — 2 sourced deals, $275,000
- Deal-DDAAF2: $180,000, UTM Source = Partner
- Deal-2CF33E: $95,000, UTM Source = Partner
- Arithmetic: $180,000 + $95,000 = $275,000
- Status: Both in DS1. Co-webinar locked for 09-15.

HRCloud Partners — 1 sourced deal, $140,000
- Deal-F1CDA5: $140,000, UTM Source = Partner
- Status: Integration referral closed the security review; opp moved to DS2.

CultureBridge — 2 sourced deals, $135,000
- Deal-096E1D: $60,000, UTM Source = Partner
- Deal-067213: $75,000, UTM Source = Partner
- Arithmetic: $60,000 + $75,000 = $135,000
- Status: Lunch-and-learn produced two sourced opps; both early stage.

WorkWell Group — 0 sourced deals, $0
- Status: No sourced deals this period. Wants to restart the joint playbook in Q4; planning call booked for 09-09.

Quiet partners

Recogniq — Quiet; no sourced deals in the data provided.
KudosWave — Quiet; no sourced deals in the data provided.
PeopleFirst Advisors — Quiet; no sourced deals in the data provided.
TotalPerk — Quiet; no sourced deals in the data provided.
rfp-answer-from-policyPASS 1.00
Plain-English question: Answer the 10 RFP items from the policy excerpts only — no guessing.
Prompt — what the model saw
Answer the 10 questionnaire items from the policy excerpts only. Cite the excerpt id for every answer. Where the excerpts do not answer the question, mark it unanswerable and say what document would; do not guess. Tone: vendor response, direct.
cost $0.0157 · 19s · in 436 / out 493 tok
✓ answered_count✓ unanswerable_flagged✓ citations✓ no_guess

✓ Correct Answer (ground truth)

Plain-English answer: Seven answered with excerpt citations (Q1-Q7), three flagged unanswerable with the document that would answer them (Q8-Q10).
The answer in one line: answered_count = 7; rule = unanswerable items must be flagged and the document that would answer them named; do not guess.
answered_count7
answer_map{"Q1": ["P1"], "Q2": ["P1"], "Q3": ["P2"], "Q4": ["P3"], "Q5": ["P4"], "Q6": ["P5"], "Q7": ["P6"]}
ruleunanswerable items must be flagged and the document that would answer them named; do not guess.
unanswerableQ8, Q9, Q10
valid_citation_idsP1, P2, P3, P4, P5, P6, P7, P8
graded checksanswered_count · unanswerable_flagged · citations · no_guess

Model's actual answer

show the model's full answer
Q1 — Yes. Customer data is encrypted at rest using AES-256. [P1]

Q2 — Yes. Data in transit is encrypted with TLS 1.2 or higher. [P1]

Q3 — Yes. Bonusly maintains a SOC 2 Type II report, renewed annually. The report is available to prospects under NDA. [P2]

Q4 — Customer data is retained for the duration of the contract plus 90 days, after which it is deleted from production systems. [P3]

Q5 — A current list of subprocessors is maintained at trust.bonusly.com. Customers receive 30 days notice before a new subprocessor is added. [P4]

Q6 — Yes. SAML 2.0 single sign-on is available on the Enterprise plan. [P5]

Q7 — Production databases are backed up daily. Backups are encrypted and retained for 35 days. [P6]

Q8 — Unanswerable from the excerpts provided. No excerpt states ISO 27001 certification status. Required document: ISO 27001 certificate, or a certifications list from the trust portal.

Q9 — Unanswerable from the excerpts provided. No excerpt states a contractual uptime SLA percentage. Required document: Master Services Agreement, Service Level Agreement, or Order Form stating the uptime commitment.

Q10 — Unanswerable from the excerpts provided. No excerpt states whether a HIPAA Business Associate Agreement will be signed. Required document: Data Processing Addendum, HIPAA/BAA policy, or Master Services Agreement.
ops-maintenance 5 tests
skill-file-reconciliationPASS 1.00
Plain-English question: Audit a skill set against its manifest and flag every conflict with a severity and an action.
Prompt — what the model saw
Reconcile the skill set below against its manifest. Find and report (with severity CRITICAL/WARNING/INFO and an action type TRIM_DESC, UPDATE_BODY, DELETE_SKILL, MERGE, or REVIEW): (1) any two skills whose ALWAYS-trigger phrases overlap/duplicate, (2) any circular delegation chain between skills (name it), (3) any delegation target that does not exist (dangling, name it), (4) any version conflict and which skill should survive, (5) how many manifest descriptions exceed 1,024 characters, (6) any hardcoded page ids, dates, or person names in skill bodies, (7) manifest drift in BOTH directions (files with no manifest row, and manifest rows with no file). Make one proposal per finding; rewrite nothing.
cost $0.3653 · 48s · in 64,387 / out 5,181 tok
✓ duplicate_trigger_found✓ circular_chain✓ dangling_target✓ oversized_desc_count✓ hardcoded_values✓ manifest_drift✓ severity_and_action

✓ Correct Answer (ground truth)

Plain-English answer: A severity-ranked list of TRIM_DESC / UPDATE_BODY / DELETE_SKILL / MERGE / REVIEW actions covering every rule violation found.
The answer in one line: files_audited = 14; rank = CRITICAL / WARNING / INFO with one fix action per issue
files_audited14
findingsduplicate trigger, circular delegation chain, dangling file reference, oversized description, hardcoded values, manifest drift
rankCRITICAL / WARNING / INFO with one fix action per issue
graded checksduplicate_trigger_found · circular_chain · dangling_target · oversized_desc_count · hardcoded_values · manifest_drift · severity_and_action

Model's actual answer

show the model's full answer
Reconciliation uses only the 14 files and the 14-row manifest provided. No skills were rewritten.

Arithmetic used below
- Manifest rows provided: 14
- Skill files provided: 14
- Description limit check: count of `description_chars > 1024`
  - Values: 656, 897, 996, 792, 965, 676, 945, 1004, 1006, 962, 1006, 708, 762, 656
  - Maximum = 1006; 1006 < 1024; count over limit = 0 / 14 = 0%

Findings

1. CRITICAL | MERGE | `comms-drafter` vs `email-drafter` — duplicate ALWAYS trigger surface
   - Evidence: `comms-drafter` triggers for any external communication and lists: “write me an email,” “draft a follow-up,” “help me reply,” “what should I say,” “bump email,” “contract nudge.”
   - `email-drafter` triggers for any customer-facing email and lists the same core phrases: “write me an email,” “draft a follow-up,” “help me reply to this,” “what should I say,” “bump email,” “contract nudge.”
   - Both also split the same boundary in opposite directions: `comms-drafter` says “For deep deal strategy, use deal-strategy-coach”; `email-drafter` says “For deal strategy… use deal-strategy-coach instead,” while each still claims the drafting lane.
   - Proposal: merge into one drafting skill, or keep one as the sole email trigger and reduce the other to non-email channels only. One proposal: MERGE.

2. CRITICAL | TRIM_DESC | `weekly-pipeline-report` vs `pipeline-intelligence-report` — duplicate pipeline trigger phrases
   - Evidence: `pipeline-intelligence-report` ALWAYS triggers for “pipeline update” and “what's the pipeline look like.”
   - `weekly-pipeline-report` ALWAYS triggers for “run the pipeline update,” “update the pipeline,” “what does pipeline look like,” and “give me this week's numbers.”
   - The overlap is on the exact user intent “pipeline update / what does pipeline look like,” but the outputs differ: 10-tab scored pipeline HTML vs weekly pipeline performance HTML.
   - Proposal: trim one description so the weekly cadence intent and the full scored-pipeline intent do not share the same trigger phrases. One proposal: TRIM_DESC.

3. WARNING | TRIM_DESC | Universal “final / always” chain among `analysis-validator`, `signalforge-claim-compressor`, and `signalforge-feedback`
   - Evidence: `analysis-validator` says “Mandatory final QA agent… Never skip”; `signalforge-claim-compressor` says “Final style pass before any SignalForge report… ALWAYS trigger”; `signalforge-feedback` says “ALWAYS trigger this skill as the absolute final step… Never skip.”
   - The bodies do define an order — data → analysis → `analysis-validator` → `signalforge-claim-compressor` → output → feedback prompt — but three descriptions each claim final/always status for the same SignalForge outputs.
   - Proposal: trim the descriptions so only one skill claims “absolute final,” and the other two state their position in the sequence instead of independent ALWAYS triggers. One proposal: TRIM_DESC.

4. WARNING | TRIM_DESC | `model-selection` vs every other skill’s start trigger
   - Evidence: `model-selection` says “ALWAYS run this skill at the start of every task, without exception — before any planning, execution, or skill invocation begins” and “before any other skill.”
   - That conflicts in practice with any skill that also says it must run first for its domain, and with the SignalForge terminal chain above, because `model-selection` claims pre-emption over all invocations.
   - Missing data: no `skill-orchestrator` file was provided, so the orchestrator-level precedence rule cannot be verified from the provided set.
   - Proposal: trim `model-selection` from “every task, without exception” to the specific planning contexts where model annotation is actually required. One proposal: TRIM_DESC.

5. CRITICAL | REVIEW | Circular delegation: `deal-strategy-coach` ↔ `email-drafter`
   - Chain: `deal-strategy-coach` → “When drafting manager-to-prospect emails, use the `email-drafter` skill.”
   - Reverse chain: `email-drafter` → “If the user needs strategic deal coaching… point them to the deal-strategy-coach skill. If they need both strategy and a draft, do the draft here and suggest they use deal-strategy-coach.”
   - A manager-email request that needs diagnosis can therefore bounce: coach sends drafting to `email-drafter`; `email-drafter` sends strategy back to `deal-strategy-coach`.
   - `comms-drafter` also points to `deal-strategy-coach`, widening the loop entry points.
   - Proposal: review and break the mutual handoff by naming one owner for manager-email drafting end-to-end. One proposal: REVIEW.

6. WARNING | REVIEW | Apparent caller/callee loop: `pipeline-intelligence-report` ↔ `closed-lost-analysis`
   - Evidence: `pipeline-intelligence-report` says “Loss Intel tab delegates to `closed-lost-analysis` skill” and Phase 2b “Delegate entirely to the `closed-lost-analysis` skill (Mode 4 — Active Pipeline Risk Match).”
   - `closed-lost-analysis` Mode 4 is explicitly triggered when “called from pipeline-intelligence-report,” and its output feeds back into pipeline scoring.
   - This is a designed return path rather than an infinite delegation in the text, but as written it is a two-skill cycle: pipeline report → closed-lost Mode 4 → pipeline report scoring/rendering.
   - Proposal: review whether Mode 4 is a sub-procedure of the pipeline report or a standalone skill, and state that ownership in one place only. One proposal: REVIEW.

7. CRITICAL | REVIEW | Dangling delegation targets not present in the provided manifest/files
   - Named targets with no manifest row and no provided file include:
     - `bonusly-brand` — invoked by `comms-drafter`, `email-drafter`, and `sales-forecast`.
     - `prospect-research-multithreading` — invoked by `comms-drafter`, `email-drafter`, and `deal-strategy-coach`.
     - `skill-orchestrator` — referenced by `analysis-validator` and `signalforge-feedback`.
     - `signalforge-reports` — referenced by `pipeline-intelligence-report` and `weekly-pipeline-report` for design files.
     - Specialist skills listed in `analysis-validator` §12.4: `bonusly-data-questions`, `bonusly-product-questions`, `bonusly-business-reporting-questions`, `bonusly-rewards-questions`, `bonusly-ppp-questions`, `bonusly-feature-flag-questions`, `bonusly-deal-desk-questions`, `bonusly-datadog-questions`.
   - Data limitation: these may exist outside the provided set; within the provided manifest they are dangling.
   - Proposal: review the manifest scope — either add rows/files for these targets or remove the delegation language from the bodies. One proposal: REVIEW.

8. WARNING | UPDATE_BODY | Version conflict in `analysis-validator`
   - Evidence: header says “Version: 3.6”; footer says “analysis-validator v3.6 · May 9, 2026”; changelog latest entry is 3.6 on May 9, 2026.
   - The validation trail template inside the same file says “Validator: analysis-validator v3.2.”
   - `pipeline-intelligence-report` footer also cites “Analysis Validator v3.6.”
   - Surviving version on the provided evidence: v3.6, because header, footer, changelog, and the cross-reference in `pipeline-intelligence-report` agree; v3.2 appears only in the trail template.
   - Proposal: update the body trail template from v3.2 to the surviving v3.6. One proposal: UPDATE_BODY.

9. INFO | REVIEW | Manifest descriptions over 1,024 characters
   - Count = 0.
   - Arithmetic: highest values are `pipeline-intelligence-report` = 1006 and `signalforge-claim-compressor` = 1006; 1024 − 1006 = 18 characters of headroom. No row exceeds 1024.
   - Proposal: no trim required on length grounds; review only if the platform limit is lower than 1,024. One proposal: REVIEW.

10. CRITICAL | UPDATE_BODY | Hardcoded page IDs, dates, and person names in skill bodies
   - Examples, cited exactly as given:
     - `partner-digest`: Cloud ID `73fe98de-a4a3-4869-9f8a-bb1eeed4cf7f`, Space ID `1958248479`, folder ID `2286616609`, page IDs `2286321666`, `2265382925`, `2236940297`, `2237825028`, `2239365136`, `2238283777`; “May 16, 2026 issue”; “Owner: Amani Phipps”; Slack search `from:<@U03QLMBL7AR>`; named contacts “Kelli, Jen Lee,” “Hani, Bryce,” “Sara.”
     - `sales-forecast`: Space ID `2232811524`, Cloud ID `73fe98de-a4a3-4869-9f8a-bb1eeed4cf7f`, Parent page ID `2232582148`; changelog names “Elena” replaced by “Alaina (VP Sales)”; example title “Q3 2026 Forecast Intelligence — July 9, 2026.”
     - `analysis-validator`: “Created: April 26, 2026,” “Last Updated: May 9, 2026,” roster with names and HubSpot owner IDs, “Escalate to Finance (Manish or Amani),” stage IDs `150582536`, `150582537`, `150582538`, `150582539`, `1175632767`.
     - `pipeline-intelligence-report`: AE owner IDs for “Bryce Harmon,” “Dana Mercer,” “Cole Ingram,” “Alex Franklin,” “Gavin Porter”; HubSpot org ID `1973303`; stage IDs `150582536`–`1175632767`.
     - `weekly-pipeline-report`: title names “Ben Lavin”; spreadsheet IDs `1CLZeOsElVDF_LF0ZG_t2nfwvhnZ6bpwqM_nX3WEYzcw` and `1ENuaEcCuLjdKhMvp8FK3Ys1ek5Aw9ZuOZhsHJJFoB_k`; static “Q1 2026 context” figures; “Q2 (April 1 – June 30, 2026; total ≈ 64–65).”
     - `signalforge-feedback`: Page ID `2295136266`, spaceId `2232811524`, Parent `2234417154`, Build Log page ID `2247295002`.
     - `stale-pipeline-report`: Slack channel ID `C0561C1JCPJ`; example dates “5/15,” “5/19.”
     - `deal-strategy-coach`: Confluence playbook page `2257879045`; “Pricing — 2026”; routed names “Perseus” and “Farid.”
     - `closed-lost-analysis`: dated/sample references including “May 2026,” “MinIO: rep vacation May 4–12,” and named account examples in the body.
   - Proposal: update bodies to move page IDs, rosters, owner IDs, spreadsheet IDs, and dated examples into referenced lookup files or runtime resolution, as `stale-pipeline-report` already requires for owner IDs (“Never hardcode rep names or owner IDs”). One proposal: UPDATE_BODY.

11. INFO | REVIEW | Manifest drift in both directions
   - Files provided with no manifest row: 0.
     - Provided files = analysis-validator, closed-lost-analysis, comms-drafter, deal-strategy-coach, email-drafter, model-selection, next-to-close, partner-digest, pipeline-intelligence-report, sales-forecast, signalforge-claim-compressor, signalforge-feedback, stale-pipeline-report, weekly-pipeline-report = 14.
     - Manifest rows = the same 14 names = 14.
     - Files − manifest rows = 14 − 14 = 0.
   - Manifest rows with no file: 0.
     - Manifest rows − files = 14 − 14 = 0.
     - Every manifest `file` value also matches its `declared_name` in the provided manifest.
   - Missing data: body-line counts and description character counts in the manifest were not independently re-measured from the full file bodies in this reconciliation, so drift on those two numeric columns is not asserted either way.
   - Proposal: no deletion on manifest-drift grounds; review only the dangling cross-skill references in Finding 7, which are references, not manifest rows. One proposal: REVIEW.

No DELETE_SKILL proposal is supported by the provided data alone: the overlaps in Findings 1–4 are resolvable by merge/trim on the evidence given, and the manifest itself is internally complete at 14 files ↔ 14 rows.
runbook-from-slack-threadPASS 1.00
Plain-English question: Turn the resolved incident thread into a runbook.
Prompt — what the model saw
Turn the resolved incident thread into a runbook: ordered steps with the exact command or action taken, who ran it, how success was verified, and rollback for each step that changed state. Every step must trace to a message in the thread. Add nothing the thread does not contain; where the thread is unclear, mark the step as needing confirmation.
cost $0.0490 · 26s · in 2,259 / out 1,586 tok
✓ ordered_steps✓ rollback✓ needs_confirmation✓ sourced

✓ Correct Answer (ground truth)

Plain-English answer: Ordered steps from the thread, each with the exact command, who ran it, how success was verified, and rollback where state changed.
The answer in one line: rule = no steps beyond what the thread contains; every step traces to a message id.
ambiguous_steporder=4; msg=M05; mark=needs-confirmation
ruleno steps beyond what the thread contains; every step traces to a message id.
stepsorder=1; msg=M01; actor=Bryce Harmon; action=acknowledge PagerDuty alert, take IC; state_change=False; verified_by=M01 | order=2; msg=M02; actor=Farid Osman; action=check queue depth; command=bundle exec rake sidekiq:queue_depth; state_change=False; verified_by=M02 (48,213 pending) | order=3; msg=M04; actor=Farid Osman; action=pause auto-enqueue via feature flag; command=bin/rails runner 'FeatureFlag.disable(:auto_recognition_enqueue)'; state_change=True; rollback=bin/rails runner 'FeatureFlag.enable(:auto_recognition_enqueue)'; verified_by=M07 (queue depth falling) | order=4; msg=M05; actor=Elena Sinclair; action=cleared the dead set; state_change=True; needs_confirmation=thread gives no command or method for clearing the dead set | order=5; msg=M06; actor=Bryce Harmon; action=scale reward-worker to 6 replicas; command=kubectl scale deployment/reward-worker --replicas=6; state_change=True; rollback=kubectl scale deployment/reward-worker --replicas=3; verified_by=M07/M08 | order=6; msg=M08; actor=Cole Ingram; action=verify queue drained and error rate baseline; command=bundle exec rake sidekiq:queue_depth; state_change=False; verified_by=M08 (returns 0) … (+2 more)
graded checksordered_steps · rollback · needs_confirmation · sourced

Model's actual answer

show the model's full answer
Runbook — Incident #incident-2026-08-29-reward-queue (resolved)

Source: incident_thread.md [M01–M10] only. Gaps are marked as needing confirmation; no step is added beyond the thread.

1. Acknowledge alert and take Incident Command — [M01]
   - Actor: Bryce Harmon
   - Action: "PagerDuty alert fired for reward-worker queue depth > 10k. Acknowledging, taking IC."
   - Exact command: None documented.
   - Success verified: No independent verification documented — needs confirmation.
   - Rollback: Not documented — needs confirmation. No state-changing command is recorded for this step in the thread.

2. Check reward queue depth — [M02]
   - Actor: Farid Osman
   - Exact command: `bundle exec rake sidekiq:queue_depth`
   - Reported result: reward queue at 48,213 pending jobs. Normal is under 500.
   - Success verified: The command return itself is the reported result in [M02].
   - Rollback: Not applicable as documented — this is a read-only check; no state change or rollback is recorded in [M02].

3. Report dead set status — [M03]
   - Actor: Farid Osman
   - Action: Reported dead set has 112 jobs, all Redis::TimeoutError from around 13:58.
   - Exact command: None documented — inspection command needs confirmation.
   - Success verified: No verification method documented beyond the report in [M03] — needs confirmation.
   - Rollback: Not applicable as documented — no state change is recorded in [M03].

4. Pause enqueue — state-changing — [M04]
   - Actor: Farid Osman
   - Exact command: `bin/rails runner 'FeatureFlag.disable(:auto_recognition_enqueue)'`
   - Reported reason: "Pausing enqueue to stop the bleed"
   - Success verified: No direct flag verification is recorded in [M04] — needs confirmation. Later queue observations in [M07]/[M08] are not recorded in the thread as direct verification of this flag change.
   - Rollback (explicitly documented in [M04]): `bin/rails runner 'FeatureFlag.enable(:auto_recognition_enqueue)'`

5. Clear dead set — state-changing — [M05]
   - Actor: Elena Sinclair
   - Action as documented: "While I was in the console I cleared out the dead set."
   - Exact command: None documented — needs confirmation.
   - Success verified: None documented in [M05] — needs confirmation.
   - Rollback: Not documented — needs confirmation. Do not mark as N/A; this is state-changing and the thread does not establish its command, verification, or rollback.
   - Note: The thread records that this happened during the incident; it does not establish it as an approved repeatable remediation.

6. Scale workers up — state-changing — [M06]
   - Actor: Bryce Harmon
   - Exact command: `kubectl scale deployment/reward-worker --replicas=6` (was 3)
   - Success verified: No direct replica-count verification is recorded in [M06] — needs confirmation. [M07] and [M08] are later queue observations, not isolated proof of this action's effect.
   - Rollback (explicitly documented in [M06]): `kubectl scale deployment/reward-worker --replicas=3`

7. Report queue draining — [M07]
   - Actor: Farid Osman
   - Action: Reported queue depth down to 9,400 and falling ~1,200/min.
   - Exact command: None documented — measurement command needs confirmation.
   - Success verified: This is a reported observation in [M07]; no separate verification is documented.
   - Rollback: Not applicable as documented — no state change is recorded in [M07].

8. Verify queue empty and error rate baseline — [M08]
   - Actor: Cole Ingram
   - Exact command: `bundle exec rake sidekiq:queue_depth`
   - Reported result: command now returns 0.
   - Success verified: Cole Ingram reports "Error rate in Datadog back to baseline" in [M08].
   - Rollback: Not applicable as documented — this is a verification step; no state change is recorded in [M08].

9. Re-enable enqueue — state-changing — [M09]
   - Actor: Bryce Harmon
   - Exact command: `bin/rails runner 'FeatureFlag.enable(:auto_recognition_enqueue)'`
   - Success verified: "Verified 40 new jobs processed cleanly in the next 3 minutes." [M09]
   - Rollback: Not documented for this step — needs confirmation. Do not infer disabling enqueue as its approved rollback.

10. Scale workers back down and resolve — state-changing — [M10]
    - Actor: Bryce Harmon
    - Exact command: `kubectl scale deployment/reward-worker --replicas=3`
    - Success verified: "Queue stable at 0. Incident resolved." [M10]
    - Rollback: Not documented for this step — needs confirmation. Do not infer scaling back to 6 as its approved rollback.
datadog-incident-timelinePASS 1.00
Plain-English question: Sequence the failure from the logs — what broke first, and what cascaded after it?
Prompt — what the model saw
From the log slice, sequence the failure: first error with timestamp, the cascade in order, the service and job involved, and the Datadog query that would confirm the first error. State what the logs do not show.
cost $0.0463 · 25s · in 1,628 / out 1,785 tok
✓ first_error_ts✓ service_named✓ cascade_ordered✓ query✓ gaps_stated

✓ Correct Answer (ground truth)

Plain-English answer: A reward-service Redis timeout at 14:01:12Z cascades to sidekiq at 14:01:40 and api-gateway at 14:03:05, with the Datadog query that confirms the first error.
The answer in one line: job_class_involved = RewardGiveJob
cascade_orderservice=reward-service; first_error_at=2026-09-03T14:01:12Z | service=sidekiq; first_error_at=2026-09-03T14:01:40Z | service=api-gateway; first_error_at=2026-09-03T14:03:05Z | service=web-app; first_error_at=2026-09-03T14:03:30Z
clean_servicespostgres
confirming_queryservice:reward-service status:error "@message:Redis::TimeoutError"
first_errortimestamp=2026-09-03T14:01:12Z; service=reward-service; message=Redis::TimeoutError: Connection to Redis at redis-primary:6379 timed out after 5s
job_class_involvedRewardGiveJob
logs_do_not_showroot cause of the Redis timeout, redis-server's own logs, infrastructure metrics (CPU/memory/network), any deploy or config change preceding the incident, who applied mitigation or when (no actor fields in logs)
graded checksfirst_error_ts · service_named · cascade_ordered · query · gaps_stated

Model's actual answer

show the model's full answer
First error
- 2026-09-03T14:01:12Z — `reward-service`, ERROR: `Redis::TimeoutError: Connection to Redis at redis-primary:6379 timed out after 5s`

Service and job involved
- Primary service: `reward-service`
- Primary job: `RewardGiveJob`
- Also affected in `sidekiq_jobs.csv`: `RecognitionDigestJob`

Cascade in order — `datadog_logs.csv`

1. 2026-09-03T14:01:12Z — `reward-service` ERROR: Redis connection timeout to `redis-primary:6379` after 5s
2. 2026-09-03T14:01:20Z — `reward-service` ERROR: `Redis::TimeoutError: retry exhausted for RewardGiveJob`
3. 2026-09-03T14:01:30Z — `reward-service` ERROR: same retry-exhausted message
4. 2026-09-03T14:01:40Z — `reward-service` ERROR: same retry-exhausted message
5. 2026-09-03T14:01:40Z — `sidekiq` ERROR: `RewardGiveJob failed: Redis::TimeoutError; retrying in 60s`
6. 2026-09-03T14:02:28Z — `sidekiq` ERROR: `RewardGiveJob failed: Redis::TimeoutError; retrying`
7. 2026-09-03T14:02:30Z — `sidekiq` WARN: `Queue reward depth above 10,000`
8. 2026-09-03T14:03:05Z — `api-gateway` ERROR: `502 upstream timeout calling reward-service /gives`
9. 2026-09-03T14:03:30Z — `web-app` ERROR: `Give form submission failed: upstream 502 from api-gateway`
10. 2026-09-03T14:03:31Z to 2026-09-03T14:06:52Z — repeated cycle in the log slice:
    - `sidekiq` ERROR: `RewardGiveJob failed: Redis::TimeoutError; retrying` at 14:03:31Z, 14:04:22Z, 14:05:26Z, 14:06:47Z
    - `api-gateway` ERROR: `502 upstream timeout calling reward-service` at 14:03:48Z, 14:04:13Z, 14:05:16Z, 14:06:52Z
    - `web-app` ERROR: `Give form submission failed: upstream 502` at 14:04:45Z, 14:05:42Z, 14:06:49Z
11. 2026-09-03T14:22:10Z — `reward-service` INFO: `Redis connection restored; resuming job processing`
12. 2026-09-03T14:24:45Z — `sidekiq` INFO: `Queue reward depth below 500`

Duration shown by the logs
- First error to Redis restoration: 14:22:10Z − 14:01:12Z = 20 minutes 58 seconds
- First error to queue below 500: 14:24:45Z − 14:01:12Z = 23 minutes 33 seconds

Failed jobs — `sidekiq_jobs.csv`
- Total rows: 16
- `RewardGiveJob`: J-00001 through J-00012 = 12 jobs
- `RecognitionDigestJob`: J-00013 through J-00016 = 4 jobs
- Arithmetic: 12 + 4 = 16
- All 16 have error `Redis::TimeoutError`
- Earliest `failed_at` in this file: 2026-09-03T14:01:46Z (J-00005, `RewardGiveJob`)

Datadog query that would confirm the first error
```
service:reward-service status:ERROR "Redis::TimeoutError: Connection to Redis at redis-primary:6379 timed out after 5s"
```
Run around 2026-09-03T14:01:12Z. Expected first matching event is the 14:01:12Z `reward-service` error above.

What the logs do not show
- Why Redis at `redis-primary:6379` timed out — no Redis service logs, Redis metrics, network, CPU/memory, failover, or configuration events are provided.
- Who or what restored the Redis connection at 14:22:10Z — no remediation action, operator, deploy, restart, or failover event is logged.
- Exact queue depth — only two thresholds are shown: `above 10,000` at 14:02:30Z and `below 500` at 14:24:45Z.
- Whether the 16 failed jobs in `sidekiq_jobs.csv` later succeeded, were retried to completion, went dead, or were lost — no eventual job outcome is provided.
- How many give submissions/users were affected — `web-app` shows give-form failures, but no counts, users, companies, or deals are provided.
- Any Postgres failure — `postgres` appears only with INFO `checkpoint complete` entries from 14:10:56Z to 14:20:59Z; no Postgres error is shown.
- Any errors between 14:06:52Z and 14:22:10Z in this slice, other than the Postgres checkpoints — the slice has no `reward-service`, `sidekiq`, `api-gateway`, or `web-app` entries in that interval.
feature-flag-state-summaryPASS 1.00
Plain-English question: Summarize what each feature flag controls and who has it on.
Prompt — what the model saw
Summarize the state of the listed feature flags: what each controls per the code excerpt, which companies or segments have it on, targeting rules, and any flag with no code reference or no targets. Company counts must match the export. Do not describe a flag that is not in the export.
cost $0.1210 · 24s · in 22,066 / out 1,433 tok
✓ orphans✓ no_targets✓ counts_match✓ no_invented

✓ Correct Answer (ground truth)

Plain-English answer: Recognition streaks on for 42 beta companies, redeem-flow redesign off with a 12-company target list, plus any flag with no code reference or no targets.
The answer in one line: rule = company counts must equal the export; do not describe flags absent from the export.
code_referencedrecognition_streaks_v2, points_budget_guardrails, slack_dm_nudges, redeem_flow_redesign, analytics_dashboard_v3, ms_teams_app_v2
flags{"recognition_streaks_v2": {"state": "on", "targeting_rules": "segment:beta_companies", "company_count": 42}, "points_budget_guardrails": {"state": "on", "targeting_rules": "all_companies", "company_c
flags_with_no_targetspaused_offboard_cleanup
orphan_flags_no_code_referencelegacy_give_modal, survey_boosters_q3, paused_offboard_cleanup
rulecompany counts must equal the export; do not describe flags absent from the export.
graded checksorphans · no_targets · counts_match · no_invented

Model's actual answer

show the model's full answer
State of listed feature flags

| flag_key | state | targeting_rules | company_count | What it controls (per code excerpt) |
|---|---|---:|---:|---|
| recognition_streaks_v2 | on | segment:beta_companies | 42 | In app/models/recognition.rb: if enabled for the company, `StreakTracker.record(give)` |
| points_budget_guardrails | on | all_companies | 220 | In app/services/budget_service.rb: if enabled for the company, `BudgetService.new(company).enforce!(giver, points)` |
| slack_dm_nudges | on | segment:region_na | 87 | In app/jobs/nudge_job.rb: job returns unless enabled for the company; otherwise `SlackDm.send_nudge(user)` |
| redeem_flow_redesign | off | targeted_list | 12 | In app/controllers/redeem_controller.rb: if enabled, `render RedeemV2Component.new(company)`; else `render RedeemV1Component.new(company)` |
| analytics_dashboard_v3 | on | segment:tier_three | 65 | In app/controllers/analytics_controller.rb: if enabled for the company, `@dashboard = AnalyticsV3.new(company)`. No fallback is shown in the excerpt. |
| ms_teams_app_v2 | off | targeted_list | 9 | In app/services/teams_installer.rb: if enabled for the company, `TeamsAppV2.install(company)` |
| legacy_give_modal | off | segment:legacy_plan | 14 | No code reference in the excerpt — what it controls is missing from the provided data |
| survey_boosters_q3 | on | segment:legacy_plan | 7 | No code reference in the excerpt — what it controls is missing from the provided data |
| paused_offboard_cleanup | off | [none — empty] | 0 | No code reference in the excerpt — what it controls is missing from the provided data |

Which companies or segments have it on

- No individual company names are provided in the export. Only segments/targeting types are given.
- Flags with state=on, by targeting: recognition_streaks_v2 — segment:beta_companies (42); points_budget_guardrails — all_companies (220); slack_dm_nudges — segment:region_na (87); analytics_dashboard_v3 — segment:tier_three (65); survey_boosters_q3 — segment:legacy_plan (7).
- Flags with state=off have targeting entries but are not on: redeem_flow_redesign — targeted_list (12); ms_teams_app_v2 — targeted_list (9); legacy_give_modal — segment:legacy_plan (14).

Flags with no code reference

legacy_give_modal, survey_boosters_q3, paused_offboard_cleanup — 3 of the 9 exported flags have no reference in flag_code.md.

Flag with no targets

paused_offboard_cleanup — targeting_rules is empty and company_count is 0. It is the only flag with no targets.

Count check (arithmetic)

- On flags: 42 + 220 + 87 + 65 + 7 = 421 flag-company assignments
- Off flags: 12 + 9 + 14 + 0 = 35 flag-company assignments
- Total across the export: 421 + 35 = 456 flag-company assignments

Note: these are assignments per flag as given in the export, not a count of unique companies — whether segments overlap is missing from the provided data.
nda-triagePASS 1.00
Plain-English question: Triage each NDA green, yellow, or red — and quote the clause that decides it.
Prompt — what the model saw
Triage each NDA as GREEN (standard approval), YELLOW (counsel review), or RED (full legal review). Cite the clause that drives each classification by quoting it. Catch embedded non-solicits, non-competes, missing carve-outs, and unusual term or governing law. One line of reasoning per NDA.
cost $0.0146 · 14s · in 524 / out 418 tok
✓ green✓ yellow✓ red✓ non_solicit_caught✓ clause_quoted✓ correct_labels

✓ Correct Answer (ground truth)

Plain-English answer: NDA-1 green (standard mutual terms), NDA-2 yellow for an embedded 18-month non-solicit, with the driving clause quoted for every classification.
The answer in one line: non_solicit_in = NDA-2
checksclassification_per_nda, clause_quotes_exist_in_document, non_solicit_caught, carveouts_checked
classifications{"NDA-1": {"classification": "GREEN", "driving_reason": "standard mutual NDA: 2-year term, standard carve-outs, Delaware law", "driving_clause_quote": "Carve-outs: information that (a) is or becomes p
non_solicit_inNDA-2
graded checksgreen · yellow · red · non_solicit_caught · clause_quoted · correct_labels

Model's actual answer

show the model's full answer
NDA-1 — GREEN: Mutual, 2-year term, complete carve-outs, and Delaware law: "information that (a) is or becomes publicly available through no breach, (b) was known prior to disclosure, (c) is received from a third party without duty of confidence, (d) is independently developed, or (e) must be disclosed by law or court order."

NDA-2 — YELLOW: Otherwise standard, but embeds an 18-month post-term non-solicit: "During the term of this Agreement and for eighteen (18) months thereafter, neither party shall solicit for employment or hire any employee of the other party with whom it came into contact under this Agreement."

NDA-3 — RED: One-way, 5-year term with no carve-outs, a 3-year non-compete, and exclusive Irish jurisdiction: "Recipient shall not, directly or indirectly, engage in or provide services to any business that competes with Discloser's business" and "the parties submit to the exclusive jurisdiction of its courts."