← back to all models ← RevenueBench home
scaledown

scaledown/summarize

Mean 0.981 · 36/40 perfect tests · $2.93 total · modalities in: text, image, file · 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-audit0.83
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 $0.2783 · 157s · in 152,120 / out 18,567 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+ = stage DS2, DS3, DS4, or DS5. Why-buys compliant = 15+ characters and not a placeholder (tbd, n/a, see notes); LONG(n) counts as compliant per instructions. Non-compliant = all other DS2+ deals (here, all had empty why_buys). Regressed = any earlier-stage timestamp (t_ds1..t_ds5) later than a later-stage timestamp. Missing DS1 = empty t_ds1.

Why-buys non-compliance (DS2+): 101 / 124 = 81.45%.
By owner: Bryce Harmon 15 / 18 = 83.33%; Dana Mercer 18 / 22 = 81.82%; Alex Franklin 48 / 61 = 78.69%; Cole Ingram 19 / 22 = 86.36%; Farid Osman 1 / 1 = 100.00%; Elena Sinclair 0 DS2+ deals, no rate computable.
No DS1 timestamp: 0 of 156.
Regressed timestamps: 0 of 156.

Process fixes: 1. Block DS2+ creation/advancement unless why-buys is 15+ substantive characters. 2. Assign each owner their non-compliant list above with a 7-day backfill deadline, then re-audit. 3. Add a weekly report flagging DS2+ deals with empty why-buys and any out-of-order stage timestamps before forecast review.

⚠️ File-mutation verifier: 1 file edit(s) FAILED this turn despite any wording above that may suggest otherwise. Run `git status` or `read_file` to confirm what actually landed.
  • `/Users/amaniphipps/.hermes/cache/scratch/deals_open.csv` — [write_file] Refusing to overwrite `/Users/amaniphipps/.hermes/cache/scratch/deals_open.csv`: `/Users/amaniphipps/.hermes/cache/scratch/deals_open.csv` exists but this task has not seen its full c…
owner-resolution0.83
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 $0.1208 · 57s · in 69,459 / out 6,737 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
Owner resolution from owners.csv:

Total open deals in deals_open.csv: 156
Deals resolved to an owner name: 156

Deals whose owner_id has no match in owners.csv: 0
List: none. All 6 distinct deal owner_ids (119337721, 84342457, 83155923, 83155924, 716654662, 701163055) match owners.csv.

Owner ids that map to archived/deactivated owners: 0 deals.
Archived owners in owners.csv (archived=true), with no open deals mapped to them:
- 1520255671, Gavin Porter
- 77260721, Hugo Lindqvist

Total pipeline amount per resolved owner:

1. Bryce Harmon, 119337721 — 35 deals — 1,054,144
24000 + 19656 + 13500 + 7000 + 2520 + 240000 + 99000 + 72000 + 70000 + 63600 + 45000 + 1 + 21000 + 23400 + 13680 + 5502 + 8160 + 1 + 11400 + 1 + 36000 + 31500 + 6000 + 10800 + 30275 + 17400 + 12600 + 18000 + 37440 + 18828 + 2880 + 36000 + 20880 + 10920 + 25200 = 1,054,144
Deals: Deal-25F752, Deal-E53952, Deal-C26D20, Deal-6787C2, Deal-A5E80A, Deal-2D1F1B, Deal-66D1FC, Deal-C6FE92, Deal-950043, Deal-D73B89, Deal-B23205, Deal-012CB1, Deal-40522D, Deal-C5658B, Deal-523604, Deal-C9C286, Deal-CA7DC0, Deal-483B2D, Deal-F0EBBB, Deal-3795AD, Deal-332637, Deal-1BEEBF, Deal-E25A09, Deal-FC22A3, Deal-036E80, Deal-BB8880, Deal-01E193, Deal-C1FA6D, Deal-7BBDFA, Deal-A62B1D, Deal-333EBB, Deal-93C8BF, Deal-1CCE5C, Deal-927338, Deal-A414F6

2. Alex Franklin, 84342457 — 67 deals — 624,310
14850 + 13770 + 11200 + 9000 + 6360 + 5400 + 3240 + 2484 + 1920 + 1080 + 7200 + 19000 + 2880 + 1400 + 4800 + 1632 + 10000 + 9300 + 2700 + 2160 + 1800 + 3600 + 3840 + 15000 + 1968 + 4000 + 3600 + 4800 + 3120 + 2520 + 9000 + 2400 + 62000 + 5400 + 5100 + 16700 + 4400 + 1620 + 2600 + 7200 + 18000 + 17000 + 8316 + 8100 + 18000 + 12600 + 24000 + 15000 + 9000 + 7200 + 3780 + 16200 + 7200 + 4680 + 1800 + 18000 + 2730 + 2400 + 3060 + 18000 + 12000 + 1800 + 4400 + 31200 + 7200 + 1600 + 60000 = 624,310
Deals: Deal-5408B0, Deal-D348E1, Deal-547B2B, Deal-403845, Deal-A2B47C, Deal-C61CF7, Deal-C6D97A, Deal-F9A08A, Deal-1FC049, Deal-BA571A, Deal-3EED2C, Deal-60C2C2, Deal-FA053A, Deal-7FA0C3, Deal-E531A6, Deal-D0BC96, Deal-5296C9, Deal-885F45, Deal-278DEC, Deal-4A13AD, Deal-8AD4A5, Deal-15D24F, Deal-9D0060, Deal-36C33F, Deal-0D0211, Deal-5AD94B, Deal-690476, Deal-6C60D4, Deal-EE195F, Deal-F436DA, Deal-034D49, Deal-6883F3, Deal-EC3025, Deal-317E6F, Deal-0D2F7A, Deal-1E2498, Deal-D1E6C2, Deal-BE3D9D, Deal-635B8E, Deal-DCA846, Deal-D9A72E, Deal-D9A12F, Deal-C2FF3C, Deal-CA5E44, Deal-4F775F, Deal-898FC5, Deal-CC08D1, Deal-792D44, Deal-293AF3, Deal-D8ABF7, Deal-46988D, Deal-E0B692, Deal-712010, Deal-13FEBD, Deal-F67D31, Deal-E73427, Deal-42F601, Deal-ED725A, Deal-55164C, Deal-B936FE, Deal-4B0BEB, Deal-D7E999, Deal-819506, Deal-530B50, Deal-3BA5EA, Deal-5FDCE4, Deal-92D97D

3. Dana Mercer, 83155923 — 24 deals — 341,195
11250 + 10500 + 9000 + 9000 + 5400 + 4800 + 4600 + 1920 + 15000 + 4200 + 18900 + 27000 + 43875 + 20000 + 60000 + 8100 + 16250 + 3150 + 5000 + 2100 + 23400 + 5400 + 7350 + 25000 = 341,195
Deals: Deal-9AAE5F, Deal-944310, Deal-B7EBD1, Deal-3974EB, Deal-2465CE, Deal-62D607, Deal-584EE5, Deal-0660B4, Deal-57887A, Deal-F336B6, Deal-215CCA, Deal-B42F46, Deal-E51FB7, Deal-9DDE86, Deal-44EA29, Deal-F40F04, Deal-5EED42, Deal-DAF1D9, Deal-87DDD1, Deal-8952F0, Deal-BA3DDC, Deal-7E2131, Deal-7599B8, Deal-F9A3C1

4. Cole Ingram, 83155924 — 22 deals — 288,161.43
2760 + 1330 + 7225.4 + 3360 + 5616 + 2700 + 7781.2 + 6947.5 + 40000 + 12168 + 2480.4 + 11116 + 11193 + 1875 + 31750 + 32175 + 4779.88 + 3334.8 + 58529.25 + 4140 + 18000 + 18900 = 288,161.43
Deals: Deal-7B3B0F, Deal-FD9F4E, Deal-AF932D, Deal-C7F9BF, Deal-8BC9F5, Deal-342E96, Deal-FF809F, Deal-A71728, Deal-B25F40, Deal-CD47A6, Deal-42326B, Deal-FA32A0, Deal-627646, Deal-E568D5, Deal-1BA595, Deal-813836, Deal-175395, Deal-2F3A66, Deal-D04904, Deal-481E24, Deal-CFE1E8, Deal-99A240

5. Farid Osman, 716654662 — 7 deals — 4,134
1249 + 1 + 1 + 1 + 2880 + 1 + 1 = 4,134
Deals: Deal-499BF6, Deal-03BA69, Deal-117863, Deal-F17780, Deal-8BA24E, Deal-2B39B0, Deal-8FDCD2

6. Elena Sinclair, 701163055 — 1 deal — 2,100
2100 = 2,100
Deal: Deal-57FF13

Check: 35 + 67 + 24 + 22 + 7 + 1 = 156 deals. 1,054,144 + 624,310 + 341,195 + 288,161.43 + 4,134 + 2,100 = 2,314,044.43 total open pipeline.
gong-hubspot-join-integrity0.75
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 $0.1313 · 76s · in 71,624 / out 9,694 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) Orphans: gong deal id with no match in deals_open

67 gong rows - 37 matched = 30 orphans.

Matched gong ids in deals_open: Deal-C6FE92, Deal-F9A3C1, Deal-5408B0, Deal-D348E1, Deal-B7EBD1, Deal-DAF1D9, Deal-C6D97A, Deal-3974EB, Deal-03BA69, Deal-944310, Deal-E53952, Deal-5EED42, Deal-333EBB, Deal-BA3DDC, Deal-AF932D, Deal-523604, Deal-C26D20, Deal-1E2498, Deal-A2B47C, Deal-3795AD, Deal-F40F04, Deal-CA7DC0, Deal-C61CF7, Deal-40522D, Deal-A414F6, Deal-D9A12F, Deal-A71728, Deal-60C2C2, Deal-547B2B, Deal-93C8BF, Deal-8952F0, Deal-8FDCD2, Deal-25F752, Deal-117863, Deal-584EE5, Deal-215CCA, Deal-4B0BEB = 37.

Remaining 30 have no match. Samples, all from gong file:
Deal-8FA85D, Deal-8FC3F9, Deal-3B7945, Deal-42B265, Deal-9CCC42

(b) Duplicate conversation keys: rows where calls_90d > distinct_conversation_keys

0 of 67. For every row calls_90d = distinct_conversation_keys, e.g. 11=11, 46=46. No excess.

(c) Open DS3+ deals with at least one logged call

deals_open total 156.
DS1 = 32, DS2 = 39.
DS1+DS2 = 71.
DS3 or later = 156 - 71 = 85.

Of the 37 matched gong rows, DS3+ = 37 - 12 = 25.
Non-DS3+ matched (12): DS1: Deal-523604, Deal-03BA69, Deal-117863, Deal-8FDCD2, Deal-A414F6 (5); DS2: Deal-3795AD, Deal-F40F04, Deal-CA7DC0, Deal-A71728, Deal-93C8BF, Deal-4B0BEB, Deal-AF932D (7).

Share = 25 / 85 = 29.4%.

Note: gong_owner_id not joined, per instruction it is a Gong user id, not a HubSpot owner id.
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 $0.0184 · 68s · in 954 / out 3,345 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
-- CANNOT BUILD CERTIFIED QUERY FROM CATALOG ALONE: no giving/recognition event table
-- with documented giver + company + timestamp columns is listed, so unique givers
-- and recognition count for a first calendar month cannot be derived without inventing a table/columns.
-- No arithmetic is shown because no verifiable event-level inputs exist in the catalog.
-- Per-company identifier column and first-month window definition are also undocumented.
-- Below selects only documented columns; it is NOT certified as the requested metrics.
SELECT
  FIRST_SUB_PAYMENT_DATE,
  M1_USERS,
  M1_GIVES,
  M1_REDEMPTIONS
-- PRODUCTION.PLG.COMPANY_COHORT_SUMMARY: only current table listing first-month aggregates (M1_*).
-- Avoided per notes: HUBSPOT_HUB_1973303.V2_LIVE.OBJECTS_DEALS (UNPOPULATED),
-- PRODUCTION.HUBSPOT.DEALS (STALE 2023-03), PRODUCTION.HUBSPOT.GONG_HUBSPOT_MAP_FAST (RETIRED).
-- No deleted-giver filter applied per business rule (must NOT be applied to historical giving counts).
-- M1_USERS/M1_GIVES NOT certified as unique givers/recognition count; M1_REDEMPTIONS NOT certified as STATE='succeeded'.
-- PRODUCTION.DEPRECATED_RECOGNITION.REDEMPTION_RECORDS_V2 is the documented STATE='succeeded' source,
-- but catalog gives no company/date key columns, so a join would require inventing columns and is omitted.
FROM PRODUCTION.PLG.COMPANY_COHORT_SUMMARY
```
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 $0.0826 · 107s · in 22,990 / out 10,599 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
CRM COMPLETENESS AUDIT — data provided only, no values invented

1. SCOPE GAP: DEALS FILE MISSING
- No deals.csv was provided.
- Cannot report completeness for owner, stage, amount, close date, why-buys.
- Cannot rank fixes by pipeline amount at stake (requires deal amount). See Section 7.

2. COMPLETENESS PERCENT PER FIELD (arithmetic shown)

Companies: n=34
- company_alias: 34/34 x100 = 100.00%, missing 0
- domain: 34/34 x100 = 100.00%, missing 0
- industry raw non-blank: 34/34 x100 = 100.00%, missing 0
  Note: normalized completeness = 0% — values use mixed taxonomy/case, see Section 5.
- employee_count: 25/34 x100 = 73.53%, missing 9
  Missing: C-EC3025, C-96039F, C-44EA29, C-D04904, C-B23205, C-60C75F, C-7BBDFA, C-50D386, C-93C8BF
  Arithmetic: 34-25=9 missing.
- hq_country: 28/34 x100 = 82.35%, missing 6
  Missing: C-2D1F1B, C-D73B89, C-44EA29, C-D04904, C-2C60E5, C-EE9FFB
  Arithmetic: 34-28=6 missing.

Contacts: n=52 rows provided, covering only 20/34 companies
- contact_key: 52/52 x100 = 100.00%
- company_alias: 52/52 x100 = 100.00%
- email raw non-blank: 52/52 x100 = 100.00%
  email valid (user@domain.tld): 48/52 x100 = 92.31%, invalid 4
- title: 39/52 x100 = 75.00%, missing 13
  Missing: CT-0000, CT-0022, CT-0072, CT-0080, CT-0081, CT-0092, CT-0120, CT-0121, CT-0122, CT-0132, CT-0141, CT-0162, CT-0170
  Arithmetic: 52-39=13.
- persona: 37/52 x100 = 71.15%, missing 15
  Missing: CT-0000, CT-0022, CT-0041, CT-0060, CT-0070, CT-0081, CT-0082, CT-0092, CT-0110, CT-0132, CT-0162, CT-0171, CT-0172, CT-0180, CT-0181
  Arithmetic: 52-37=15.
- domain (company domain column): 52/52 x100 = 100.00%
- Missing both title+persona (6): CT-0000, CT-0022, CT-0081, CT-0092, CT-0132, CT-0162
- Companies with zero contacts in extract (14): C-2C60E5, C-B97B4E, C-2D7423, C-7BBDFA, C-50D386, C-BA969B, C-332637, C-93C8BF, C-EE9FFB, C-C9BB20, C-0A092931, C-0A092932, C-0A092933, C-0A092934

3. DUPLICATE COMPANY CLUSTERS (shared domain, exact lowercase match)
No other name-variant clusters detectable — aliases are coded C-XXXXXX.

Cluster 1 — domain acme-corp.com
- Members: C-0A092931 [Technology, 500, US], C-0A092932 [tech, 510, USA]
- Survivor recommended: C-0A092931
  Reason: cleaner casing (Technology vs tech); no enrichment row to arbitrate 500 vs 510, requires source validation before merge. Normalize hq to chosen standard (US vs USA).

Cluster 2 — domain globex.io
- Members: C-0A092933 [SaaS, 200, US], C-0A092934 [Technology, 200, US]
- Survivor recommended: C-0A092934
  Reason: Technology matches dominant CRM taxonomy; preserve SaaS label as note. Requires business rule confirmation — do not delete loser until deal/contact reassignment is confirmed. No enrichment row for globex.io.

Fix: merge duplicates, re-point contacts/deals to survivor, then delete loser.

4. INVALID EMAILS AND DOMAIN MISMATCHES

Invalid (no domain part after @, fails user@domain.tld) — 4:
- CT-0010 (C-66D1FC): user0@
- CT-0080 (C-92D97D): user0@
- CT-0081 (C-92D97D): user1@
- CT-0192 (C-425E2A): user2@

Domain mismatches (email domain vs contacts.domain column) — 5 rows:
- CT-0010: email domain '' vs 66d1fc.com (invalid, above)
- CT-0011 (C-66D1FC): user1@other-domain.com vs 66d1fc.com — only valid-format mismatch, possible personal/old domain
- CT-0080: '' vs 92d97d.com
- CT-0081: '' vs 92d97d.com
- CT-0192: '' vs 425e2a.com

Fix: re-collect 4 invalid; verify CT-0011 ownership / replace with corporate domain or confirm opt-in.

5. CRM vs ENRICHMENT — FILL AND DISAGREEMENTS

Enrichment covers 25/34 company rows (25 domains). No match for 9 rows (5 unique domains + duplicates):
- No ZI row: C-BA969B, C-332637, C-93C8BF, C-EE9FFB, C-C9BB20, C-0A092931, C-0A092932, C-0A092933, C-0A092934
- Cannot fill or arbitrate these from enrichment.

Fills allowed only where CRM blank and ZI non-blank:
- C-EC3025: emp '' -> 400 (ZI ec3025.com). hq already USA, no overwrite.
- C-96039F: emp '' -> 400 (ZI 96039f.com)
- C-44EA29: emp '' -> 400 (ZI 44ea29.com). hq '' remains missing (ZI also blank).
- C-D04904: emp '' -> 400 (ZI d04904.com). hq '' remains missing (ZI also blank).
- C-B23205: emp '' -> 400 (ZI b23205.com)
- C-60C75F: emp '' -> 400 (ZI 60c75f.com)
- C-7BBDFA: emp '' -> 400 (ZI 7bbdfa.com)
- C-50D386: emp '' -> 400 (ZI 50d386.com)
- No fill possible: C-2D1F1B (ZI hq blank), C-D73B89 (ZI hq blank), C-2C60E5 (ZI hq blank), C-93C8BF (no ZI), C-EE9FFB hq (no ZI)

Disagreements (CRM | ZI — recommend source):
Exact matches (no action): C-C6FE92, C-31ED2A, C-63A874, C-B97B4E agree on all 3 fields.

All others differ in format/taxonomy — do not auto-overwrite, choose one standard:
- hq format US vs USA vs United States, e.g.:
  C-66D1FC: US | United States
  C-950043: US | United States
  C-EC3025: USA | United States
  C-96039F: USA | United States
  C-77A95A: US | United States
  C-B23205: US | United States
  C-E51FB7: USA | United States
  C-D0662E: US | United States
  C-425E2A: USA | United States
  C-2D7423: USA | United States
  Recommendation: CRM is system of record; normalize all to one chosen form (e.g. ISO US/Canada/UK) — never invented here.
- industry taxonomy/case, e.g.:
  C-66D1FC: tech | Computer Software
  C-EC3025: Technology | Computer Software
  C-44EA29: tech | Computer Software
  C-92D97D: Technology | Computer Software
  C-D04904: Technology | Computer Software
  C-77A95A: Technology | Computer Software
  C-AA8DDA: Technology | Computer Software
  C-60C75F: tech | Computer Software
  C-425E2A: Tech [trailing space] | Computer Software
  C-B25F40: Technology | Computer Software
  C-7BBDFA / C-50D386: health care | health care (agree, but non-standard case)
  Recommendation: define canonical industry list, map tech/Tech /Technology/Computer Software/SaaS/health care to it; keep ZI Computer Software only as reference until standard is chosen.
- employee_count conflicts where both filled: none — all filled CRM emp values equal ZI emp where both present (e.g. C-66D1FC 900=900, C-B97B4E 1500=1500). Only blank-vs-filled cases above.

6. OTHER NORMALIZATION FIXES (no invention)
- Trim trailing spaces: C-425E2A, C-BA969B, C-C9BB20 industry Tech with trailing space.
- Lowercase variants: tech (C-66D1FC, C-44EA29, C-60C75F, C-0A092932), health care (C-7BBDFA, C-50D386).

7. TOP 10 FIXES BY PIPELINE AMOUNT AT STAKE
Cannot produce — data missing. No deals file with amount was provided, so amount-at-stake arithmetic is impossible. To produce this, provide deals.csv with company_alias + amount.

Unranked highest-impact data fixes (no amount implied):
1. Resolve acme-corp.com duplicate (C-0A092931 vs C-0A092932, emp 500 vs 510)
2. Resolve globex.io duplicate (C-0A092933 SaaS vs C-0A092934 Technology)
3. Re-collect invalid emails CT-0010, CT-0080, CT-0081, CT-0192
4. Verify CT-0011 other-domain.com
5. Fill 8 missing emp values from ZI (list in Section 5)
6. Collect hq for C-2D1F1B, C-D73B89, C-44EA29, C-D04904, C-2C60E5, C-EE9FFB (no ZI fill available)
7. Collect emp for C-93C8BF (no ZI)
8. Backfill titles for 13 contacts + personas for 15 contacts (Section 2)
9. Add contacts for 14 companies with zero coverage
10. Normalize industry + hq_country to single standard
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 $0.0829 · 159s · in 24,135 / out 12,398 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
Rule used for side: buyer = text shows a buyer decision, plan, or preference; Bonusly = text cites a Bonusly fit gap (feature, UI, geo, capability); unknown = vague, single-word, spam, or ghosted with no actor.

Classification (deal_alias | category | side):

Deal-DB0AAC | timing | buyer
Deal-F7F635 | competitor | buyer
Deal-AC944F | no decision | unknown
Deal-214060 | no decision | unknown
Deal-91A056 | timing | buyer
Deal-29326C | timing | unknown
Deal-5DB9B0 | other | unknown
Deal-831B7B | timing | buyer
Deal-F97C37 | competitor | buyer
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 | buyer
Deal-ED9AE7 | other | unknown
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 | buyer
Deal-ACE061 | competitor | buyer
Deal-BB78F3 | timing | buyer
Deal-D48E0B | no decision | unknown
Deal-15DA99 | timing | buyer
Deal-F4AF5D | timing | buyer
Deal-79B7A1 | timing | unknown
Deal-583ADB | no decision | unknown
Deal-8E27DA | no decision | buyer
Deal-2D2F8D | competitor | unknown
Deal-E0441F | no decision | unknown
Deal-7CB44D | no decision | unknown
Deal-0F96AA | competitor | buyer
Deal-1BCA50 | competitor | buyer
Deal-7CC678 | competitor | unknown
Deal-FAC17C | other | buyer
Deal-242273 | competitor | buyer
Deal-50E5D8 | timing | buyer
Deal-A2C349 | competitor | buyer
Deal-9F176A | timing | buyer
Deal-7B2236 | pricing | buyer
Deal-AFA56C | no decision | unknown
Deal-C7156E | competitor | buyer
Deal-C33D91 | pricing | buyer
Deal-9048EB | product gap | Bonusly
Deal-5E64CE | competitor | buyer
Deal-8A0992 | competitor | buyer
Deal-D0C698 | competitor | buyer
Deal-69CF3D | timing | unknown
Deal-ECBF89 | timing | unknown
Deal-3618CC | product gap | Bonusly
Deal-EECC02 | competitor | unknown
Deal-5AD03E | product gap | Bonusly
Deal-D1A623 | timing | unknown
Deal-413C56 | no decision | buyer
Deal-47F1A1 | competitor | buyer
Deal-BF2A98 | competitor | buyer
Deal-2A292B | no decision | buyer
Deal-D1AABF | no decision | unknown
Deal-FEDBCB | timing | buyer
Deal-1E7DA9 | competitor | buyer
Deal-2BBA21 | no decision | unknown
Deal-286F9C | competitor | buyer
Deal-7FBAC6 | timing | buyer
Deal-369281 | competitor | buyer
Deal-386F6E | no decision | unknown
Deal-9FCD0D | competitor | buyer
Deal-55867E | timing | buyer
Deal-DAFB82 | pricing | buyer
Deal-2FEDDB | timing | buyer
Deal-64B19A | competitor | buyer
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 | buyer
Deal-5885B9 | no decision | unknown

Summary arithmetic:

Categories: no decision 28 + competitor 26 + timing 23 + pricing 5 + product gap 4 + other 3 + champion left 1 = 90.
Sides: buyer 52 + unknown 34 + Bonusly 4 = 90.

Tag-vs-text clear disagreements: 5 deals.
Deal-8E27DA (tag Feature Request; text: took swag only, did not want R&R currently)
Deal-9048EB (tag MIA; text: bad fit, desired setup, multiple feature gaps)
Deal-5E64CE (tag Doing nothing/Not a priority/Cost; text: locked with Nectar through Oct 2027, exit fee, will revisit to move to Bonusly)
Deal-3618CC (tag Lost DM; text: Wanted Surveys)
Deal-5AD03E (tag Competitor; text: wanted more defined budget access, no vendor named)

Two patterns worth acting on:

1. Dated 2027/2028 pause pile inside timing (23). Explicit early-2027 reconnects: Deal-91A056, Deal-B6AC09, Deal-E6E80A, Deal-B038F0, Deal-175756, Deal-15DA99; 2028/budget dates: Deal-B3ABED (Q2 next year for 2028 budget), Deal-DAFB82 (not budgeted until 2028); plus end-of-year: Deal-9F176A, Deal-FEDBCB. Act: dated nurture with owner and re-open dates, not dead pipeline.

2. Post-intro/demo ghosting inside no decision (28) and unknown side (34). Repeated wording: no contact since intro, ignored outreach/nudges: Deal-F308CA, Deal-4664E1, Deal-7CB44D, Deal-2BBA21, Deal-096750; single-word-real-content rows: Deal-AC944F, Deal-214060, Deal-21B045, Deal-D48E0B, Deal-583ADB, Deal-5885B9, Deal-79E61A, Deal-AE7C4E, Deal-DAB4F1, Deal-B4B50F. Act: tighten intro-to-next-step commitment and stale-deal exit rules.
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 $0.1774 · 88s · in 112,952 / out 8,290 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": 3, "RISKY": 8, "ACTION": 10, "BUILD": 41, "REVIVE": 30, "WATCH": 64}, "tier_examples": {"LOCK": ["Deal-D348E1", "Deal-C26D20", "Deal-403845"], "RISKY": ["Deal-547B2B", "Deal-B7EBD1", "Deal-A2B47C"], "ACTION": ["Deal-25F752", "Deal-C6FE92", "Deal-944310"], "BUILD": ["Deal-93C8BF", "Deal-03BA69", "Deal-117863"], "REVIVE": ["Deal-E53952", "Deal-5408B0", "Deal-9AAE5F"], "WATCH": ["Deal-2D1F1B", "Deal-950043", "Deal-B23205"]}, "risky_deals": ["Deal-547B2B", "Deal-B7EBD1", "Deal-A2B47C", "Deal-2465CE", "Deal-584EE5", "Deal-FD9F4E", "Deal-A5E80A", "Deal-499BF6"], "lock_violations": 0, "pipeline_shape": "156 deals = 3 LOCK + 8 RISKY + 10 ACTION + 41 BUILD + 30 REVIVE + 64 WATCH (11 COMMIT + 40 BEST_CASE + 105 PIPELINE); pipeline is bottom-heavy with 105 PIPELINE vs 11 COMMIT, closable core is tiny (3 LOCK), 8 COMMIT are RISKY with zero meetings or DS1/DS2 stage, and 30 BEST_CASE have zero meetings_30d."}
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 $0.0182 · 57s · in 1,786 / out 3,066 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
No arithmetic required — all dollar figures and dates below are quoted verbatim from prospect lines; no derived numbers.

```json
[
  {
    "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 it manually", "track everything in a spreadsheet, and people slip through the cracks"],
    "stakeholders": ["Prospect (VP People)", "Prospect (HR Admin)"],
    "budget_signal": "We have about $40k earmarked for engagement tools this fiscal year.",
    "timeline_signal": "Ideally we would have this live before open enrollment in November.",
    "competitor_mentioned": "Achievers",
    "competitor_context": "We looked at Achievers last year, but it was too heavy for a team our size.",
    "next_step": "Security review on September 12",
    "next_step_agreed": true,
    "objections": ["we 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 our hourly workforce"],
    "pain_points": ["regretted turnover there is over 30% for hourly workforce", "Integration with Workday has to be rock solid"],
    "stakeholders": ["Prospect (Head of Total Rewards)", "Prospect (CFO)"],
    "budget_signal": "Finance has approved a $25k pilot budget for this quarter.",
    "timeline_signal": "We want a decision by end of September.",
    "competitor_mentioned": null,
    "competitor_context": "Prospect stated: You're the first vendor we've had a real demo with.",
    "next_step": "Send the pilot agreement and we'll route it to legal this week",
    "next_step_agreed": true,
    "objections": ["Integration with Workday has to be rock solid — that's my one condition"],
    "confidence": "high"
  },
  {
    "transcript_id": "TX-003",
    "deal_alias": "Deal-530B50",
    "why_buys": ["make recognition visible across our 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,
    "budget_note": "No prospect-stated budget; $8 per employee per month was stated by Alex Franklin only — excluded per rule.",
    "timeline_signal": "Honestly there's no rush on our side until Q1.",
    "competitor_mentioned": "Bucketlist",
    "competitor_context": "My CEO used Bucketlist at her last company and liked it.",
    "next_step": "Schedule a call with our CEO — prospect to send two times",
    "next_step_agreed": true,
    "objections": ["The CEO has to be sold first — she decides anything people-related"],
    "confidence": "low"
  },
  {
    "transcript_id": "TX-004",
    "deal_alias": "Deal-180D02",
    "why_buys": ["consolidate three separate recognition tools into one"],
    "pain_points": ["We're paying for three tools and none of them talk to our HRIS", "procurement cycle runs six to eight weeks minimum", "security review took three months for our last vendor"],
    "stakeholders": ["Prospect (VP People)", "Prospect (IT Security Lead)"],
    "budget_signal": "If it's under $15k annually, I can approve it without going to the board.",
    "timeline_signal": "Our procurement cycle runs six to eight weeks minimum.",
    "competitor_mentioned": null,
    "next_step": null,
    "next_step_note": "No explicitly agreed next step; prospect said Maybe — I need to check her calendar, no promises.",
    "objections": ["The security review took three months for our last vendor — that's my hesitation"],
    "confidence": "low"
  },
  {
    "transcript_id": "TX-005",
    "deal_alias": "Deal-F8767A",
    "why_buys": ["automate service milestones", "give us analytics on recognition equity across departments"],
    "pain_points": ["Our night-shift teams feel invisible — their engagement scores run 20 points lower", "Our exec team is skeptical after a failed rollout two years ago"],
    "stakeholders": ["Prospect (HR Director)", "Prospect (People Ops Coordinator)"],
    "budget_signal": "We have $12k approved under our engagement line.",
    "timeline_signal": "We need this running before our January all-hands.",
    "competitor_mentioned": "Nectar",
    "competitor_context": "We're mid-pilot with Nectar right now, so you'd need to beat that experience.",
    "next_step": "Come present to our exec team on October 2",
    "next_step_agreed": true,
    "objections": ["Our exec team is skeptical after a failed rollout two years ago", "you'd need to beat that [Nectar] experience"],
    "confidence": "medium"
  },
  {
    "transcript_id": "TX-006",
    "deal_alias": "Deal-EE195F",
    "why_buys": ["cut the admin time on service awards"],
    "pain_points": ["I personally spend five hours a month ordering and shipping plaques", "Budget isn't the issue — time is", "our COO usually prefers building things in-house"],
    "stakeholders": ["Prospect (HR Manager)"],
    "budget_signal": null,
    "budget_note": "No prospect-stated amount; prospect stated Budget isn't the issue — time is.",
    "timeline_signal": "Our fiscal year resets in February, so a Q1 start is realistic.",
    "competitor_mentioned": null,
    "competitor_note": "Prospect stated comparing against just doing it internally / Nobody else — no named vendor raised by prospect.",
    "next_step": "Send the one-page overview and I'll forward it to our COO this week",
    "next_step_agreed": true,
    "objections": ["Fair warning, our COO usually prefers building things in-house"],
    "confidence": "medium"
  }
]
```
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 $0.0091 · 51s · in 1,101 / out 1,128 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
1. Deal-547B2B — $11,200
Why close: CRM DS5/COMMIT, close 2026-09-11. Slack 09:12 Alex Franklin: redlines clean, signing page out, VP People signing tomorrow. Signature-imminent.
Left: signature only.

2. Deal-403845 — $9,000
Why close: CRM DS5/COMMIT, close 2026-09-11. Slack 10:02 Dana Mercer: order form with their finance team.
Left: finance approval + signature.

3. Deal-B7EBD1 — $9,000
Why close: CRM DS5/COMMIT, close 2026-09-10. No contrary Slack update.
Left: final signature; specific remaining steps missing in data.

Excluded: Deal-2465CE ($5,400, DS5/COMMIT) — Slack 09:20 Dana Mercer: champion left, procurement froze vendors, pulling from commit, blocked to 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 $0.0311 · 44s · in 20,298 / out 1,342 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 (5 excerpts):

1. Deal-EC3025 — Prospect (IT Security Lead):
Quote: "We need SCIM user provisioning — our security team will not approve any tool that requires manual account management."
Classification: real gap. Product docs: "SCIM user provisioning and ADP Workforce Now integrations are NOT currently listed as supported."
Amount: not provided in data.

2. Deal-D0D6B5 — Prospect (HRIS Manager):
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."
Classification: real gap. Product docs: HRIS supported = Workday, BambooHR, Gusto, Rippling only; "ADP Workforce Now integrations are NOT currently listed."
Amount: not provided in data.

3. Deal-CFE7F4 — Prospect (VP People):
Quote: "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, not a gap. Product docs: Custom report builder = Enterprise only.
Amount: not provided in data.

4. Deal-84DBA6 — Prospect (People Ops Manager):
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."
Classification: rollout/enablement issue. Product docs: Slack integration = yes on all plans.
Amount: not provided in data.

5. Deal-36C33F — Excluded. No prospect-raised gap. The mobile-app line is rep voice: Alex Franklin: "Full transparency — we don't have a native mobile app for hourly workers yet, though it's on the roadmap." Prospect reply: "Good to know. The web version should be fine for our office staff for now." Rep language does not count.

Arithmetic: 5 excerpts reviewed = 2 real gaps + 1 plan gate + 1 rollout/enablement + 1 rep-raised exclusion. Deal amounts missing for all deals, so no pipeline total can be calculated.

Summary — true product gaps only:
- Deal-EC3025 — SCIM user provisioning missing. Amount: not provided.
- Deal-D0D6B5 — ADP Workforce Now integration missing. Amount: not provided.
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 $0.2621 · 117s · in 122,974 / out 24,679 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. Stale = max(last_email,last_call,last_meeting) in engagements_by_deal_90d.csv is >7 days before snapshot, i.e. on/before 2026-08-28. Days = 2026-09-05 minus that max date. Deal last_contacted_field ignored.

2 open deals have no row in engagements table: Deal-3EED2C, Deal-57FF13 — treated as stale with unknown recency.

Bryce Harmon — 13 stale, total 626,243
626,243 = 240000+99000+70000+45000+37440+36000+31500+23400+21000+11400+6000+5502+1
- Deal-2D1F1B, DS1, 240000, last 2026-06-16, 81 days. 09-05 minus 06-16 = 81
- Deal-66D1FC, DS1, 99000, last 2026-08-20, 16 days. 09-05 minus 08-20 = 16
- Deal-950043, DS1, 70000, last 2026-08-17, 19 days. 09-05 minus 08-17 = 19
- Deal-B23205, DS1, 45000, last 2026-08-20, 16 days. 09-05 minus 08-20 = 16
- Deal-7BBDFA, DS3, 37440, last 2026-07-21, 46 days. 09-05 minus 07-21 = 46
- Deal-332637, DS2, 36000, last 2026-08-27, 9 days. 09-05 minus 08-27 = 9
- Deal-1BEEBF, DS1, 31500, last 2026-08-17, 19 days. 09-05 minus 08-17 = 19
- Deal-C5658B, DS1, 23400, last 2026-08-20, 16 days. 09-05 minus 08-20 = 16
- Deal-40522D, DS3, 21000, last 2026-08-17, 19 days. 09-05 minus 08-17 = 19
- Deal-F0EBBB, DS3, 11400, last 2026-08-12, 24 days. 09-05 minus 08-12 = 24
- Deal-E25A09, DS1, 6000, last 2026-08-27, 9 days. 09-05 minus 08-27 = 9
- Deal-C9C286, DS2, 5502, last 2026-08-27, 9 days. 09-05 minus 08-27 = 9
- Deal-012CB1, DS1, 1, last 2026-08-13, 23 days. 09-05 minus 08-13 = 23

Dana Mercer — 14 stale, total 261,645
261,645 = 60000+43875+27000+23400+20000+18900+16250+15000+9000+9000+8100+5000+4200+1920
- Deal-44EA29, DS2, 60000, last 2026-08-26, 10 days. 09-05 minus 08-26 = 10
- Deal-E51FB7, DS2, 43875, last 2026-08-24 email/call max 2026-08-24, 12 days. 09-05 minus 08-24 = 12
- Deal-B42F46, DS1, 27000, last 2026-08-17, 19 days. 09-05 minus 08-17 = 19
- Deal-BA3DDC, DS3, 23400, last 2026-08-21 call, 15 days. 09-05 minus 08-21 = 15
- Deal-9DDE86, DS2, 20000, last 2026-08-21, 15 days. 09-05 minus 08-21 = 15
- Deal-215CCA, DS3, 18900, last 2026-08-19 meeting, 17 days. 09-05 minus 08-19 = 17
- Deal-5EED42, DS3, 16250, last 2026-08-25 email/call, 11 days. 09-05 minus 08-25 = 11
- Deal-57887A, DS2, 15000, last 2026-08-28, 8 days. 09-05 minus 08-28 = 8
- Deal-B7EBD1, DS5, 9000, last 2026-08-20 email, 16 days. 09-05 minus 08-20 = 16
- Deal-3974EB, DS4, 9000, last 2026-08-28 email/meeting, 8 days. 09-05 minus 08-28 = 8
- Deal-F40F04, DS2, 8100, last 2026-08-21 email/meeting, 15 days. 09-05 minus 08-21 = 15
- Deal-87DDD1, DS1, 5000, last 2026-08-17, 19 days. 09-05 minus 08-17 = 19
- Deal-F336B6, DS3, 4200, last 2026-08-21, 15 days. 09-05 minus 08-21 = 15
- Deal-0660B4, DS4, 1920, last 2026-08-20 meeting, 16 days. 09-05 minus 08-20 = 16

Cole Ingram — 18 stale, total 252,905.03
252,905.03 = 58529.25+40000+32175+31750+18000+12168+11193+7781.20+7225.40+6947.50+5616+4779.88+4140+3360+3334.80+2700+1875+1330
- Deal-D04904, DS2, 58529.25, last 2026-08-25, 11 days. 09-05 minus 08-25 = 11
- Deal-B25F40, DS3, 40000, last 2026-08-28, 8 days. 09-05 minus 08-28 = 8
- Deal-813836, DS2, 32175, last 2026-08-25, 11 days. 09-05 minus 08-25 = 11
- Deal-1BA595, DS2, 31750, last 2026-08-25, 11 days. 09-05 minus 08-25 = 11
- Deal-CFE1E8, DS3, 18000, last 2026-08-25, 11 days. 09-05 minus 08-25 = 11
- Deal-CD47A6, DS2, 12168, last 2026-08-25 email/call, 11 days. 09-05 minus 08-25 = 11
- Deal-627646, DS3, 11193, last 2026-08-25, 11 days. 09-05 minus 08-25 = 11
- Deal-FF809F, DS2, 7781.2, last 2026-08-25, 11 days. 09-05 minus 08-25 = 11
- Deal-AF932D, DS2, 7225.4, last 2026-08-25, 11 days. 09-05 minus 08-25 = 11
- Deal-A71728, DS2, 6947.5, last 2026-08-25, 11 days. 09-05 minus 08-25 = 11
- Deal-8BC9F5, DS2, 5616, last 2026-08-26, 10 days. 09-05 minus 08-26 = 10
- Deal-175395, DS3, 4779.88, last 2026-08-25, 11 days. 09-05 minus 08-25 = 11
- Deal-481E24, DS3, 4140, last 2026-08-26 call, 10 days. 09-05 minus 08-26 = 10
- Deal-C7F9BF, DS2, 3360, last 2026-08-25 email/call, 11 days. 09-05 minus 08-25 = 11
- Deal-2F3A66, DS3, 3334.8, last 2026-08-25, 11 days. 09-05 minus 08-25 = 11
- Deal-342E96, DS2, 2700, last 2026-08-12, 24 days. 09-05 minus 08-12 = 24
- Deal-E568D5, DS3, 1875, last 2026-08-25, 11 days. 09-05 minus 08-25 = 11
- Deal-FD9F4E, DS5, 1330, last 2026-08-26, 10 days. 09-05 minus 08-26 = 10

Alex Franklin — 19 stale, total 109,536
109,536 = 24000+18000+9300+8316+7200+5100+4800+4680+3840+3600+3240+3120+2700+2600+2400+2160+1800+1600+1080
- Deal-CC08D1, DS1, 24000, last 2026-08-20, 16 days. 09-05 minus 08-20 = 16
- Deal-E73427, DS3, 18000, last 2026-08-26 email/meeting, 10 days. 09-05 minus 08-26 = 10
- Deal-885F45, DS2, 9300, last 2026-08-24, 12 days. 09-05 minus 08-24 = 12
- Deal-C2FF3C, DS1, 8316, last 2026-08-26, 10 days. 09-05 minus 08-26 = 10
- Deal-3EED2C, DS2, 7200, no engagements row — recency missing, days unknown
- Deal-0D2F7A, DS3, 5100, last 2026-08-24 call, 12 days. 09-05 minus 08-24 = 12
- Deal-6C60D4, DS3, 4800, last 2026-08-24 call, 12 days. 09-05 minus 08-24 = 12
- Deal-13FEBD, DS2, 4680, last 2026-08-24 call, 12 days. 09-05 minus 08-24 = 12
- Deal-9D0060, DS3, 3840, last 2026-08-24, 12 days. 09-05 minus 08-24 = 12
- Deal-690476, DS2, 3600, last 2026-08-18 call, 18 days. 09-05 minus 08-18 = 18
- Deal-C6D97A, DS4, 3240, last 2026-08-28 email, 8 days. 09-05 minus 08-28 = 8
- Deal-EE195F, DS3, 3120, last 2026-08-28, 8 days. 09-05 minus 08-28 = 8
- Deal-278DEC, DS3, 2700, last 2026-08-28, 8 days. 09-05 minus 08-28 = 8
- Deal-635B8E, DS3, 2600, last 2026-08-18, 18 days. 09-05 minus 08-18 = 18
- Deal-6883F3, DS1, 2400, last 2026-08-20 email/meeting, 16 days. 09-05 minus 08-20 = 16
- Deal-4A13AD, DS3, 2160, last 2026-08-10, 26 days. 09-05 minus 08-10 = 26
- Deal-F67D31, DS2, 1800, last 2026-08-28, 8 days. 09-05 minus 08-28 = 8
- Deal-5FDCE4, DS3, 1600, last 2026-08-24, 12 days. 09-05 minus 08-24 = 12
- Deal-BA571A, DS4, 1080, last 2026-08-18, 18 days. 09-05 minus 08-18 = 18

Elena Sinclair — 1 stale, total 2,100
- Deal-57FF13, DS1, 2100, no engagements row — recency missing, days unknown

Farid Osman — 0 stale, total 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 $0.1984 · 113s · in 91,586 / out 19,744 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. DS2 window: t_ds2 from 2026-08-06 to 2026-09-05 inclusive (30 days).

Data gap: 2 deals in deals_open.csv have no row in engagements_by_deal_90d.csv and are excluded from activity totals: Deal-3EED2C (owner Alex Franklin), Deal-57FF13 (owner Elena Sinclair). Elena Sinclair has no engagement rows at all.

Per-rep last-30-day activities (sum of emails_30d + calls_30d + meetings_30d over that owner's deals):

Bryce Harmon (119337721): emails 162, calls 0, meetings 43. Total = 162+0+43 = 205.
Dana Mercer (83155923): emails 84, calls 18, meetings 11. Total = 84+18+11 = 113.
Alex Franklin (84342457): emails 307, calls 36, meetings 41. Total = 307+36+41 = 384.
Cole Ingram (83155924): emails 96, calls 14, meetings 1. Total = 96+14+1 = 111.
Farid Osman (716654662): emails 38, calls 0, meetings 34. Total = 38+0+34 = 72.
Elena Sinclair (701163055): emails 0, calls 0, meetings 0 in provided engagement data (only deal Deal-57FF13 has no engagement row). Total = 0.

Deals entering DS2 in window (t_ds2 in window):

Bryce Harmon: 4 — Deal-1CCE5C (2026-08-06), Deal-25F752 (2026-08-10), Deal-CA7DC0 (2026-08-12), Deal-D73B89 (2026-09-03).
Dana Mercer: 1 — Deal-57887A (2026-08-07).
Alex Franklin: 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-92D97D (2026-09-02), Deal-403845 (2026-09-02), Deal-1FC049 (2026-09-03), Deal-3EED2C (2026-09-03).
Cole Ingram: 2 — Deal-1BA595 (2026-08-12), Deal-42326B (2026-08-26).
Farid Osman: 1 — Deal-499BF6 (2026-08-26).
Elena Sinclair: 0.

Activity mix (share of that rep's total) and efficiency (total activities / DS2 entries):

Alex Franklin: emails 307/384 = 79.9%, calls 36/384 = 9.4%, meetings 41/384 = 10.7%. Efficiency = 384/18 = 21.3 activities per DS2 entry.
Bryce Harmon: emails 162/205 = 79.0%, calls 0/205 = 0.0%, meetings 43/205 = 21.0%. Efficiency = 205/4 = 51.3 per DS2 entry.
Cole Ingram: emails 96/111 = 86.5%, calls 14/111 = 12.6%, meetings 1/111 = 0.9%. Efficiency = 111/2 = 55.5 per DS2 entry.
Farid Osman: emails 38/72 = 52.8%, calls 0/72 = 0.0%, meetings 34/72 = 47.2%. Efficiency = 72/1 = 72.0 per DS2 entry.
Dana Mercer: emails 84/113 = 74.3%, calls 18/113 = 15.9%, meetings 11/113 = 9.7%. Efficiency = 113/1 = 113.0 per DS2 entry.
Elena Sinclair: no recorded activities; 0 DS2 entries; activities per DS2 entry not defined.

Rank by efficiency (lowest activities per DS2 entry = most efficient):
1. Alex Franklin — 21.3
2. Bryce Harmon — 51.3
3. Cole Ingram — 55.5
4. Farid Osman — 72.0
5. Dana Mercer — 113.0
Elena Sinclair — not rankable (0/0).

Most efficient rep: Alex Franklin (21.3 activities per DS2 entry).
Highest-volume rep: Alex Franklin (384 total activities).
They do not differ; they are the same 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 $0.1793 · 109s · in 99,229 / out 10,732 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
Alex Franklin — QTD Scorecard as of 2026-09-05 (2026-Q3, quota snapshot 2026-09-05)

1. Bookings vs quota
Quota: $200,000
QTD bookings (closed-won, close_date 2026-07-01 to 2026-09-05): $150,000
Arithmetic: 40,000 (Deal-A1C3E5) + 20,000 (Deal-F2C7D8) + 35,000 (Deal-B7D2F4) + 21,000 (Deal-C9E1A6) + 12,000 (Deal-A8B4D6) + 11,000 (Deal-D4B8C2) + 6,500 (Deal-E6F3A9) + 4,500 (Deal-C5D9E2) = 150,000
Excluded pre-quarter won: Deal-B3E6F1, $24,000, 2026-06-20
Attainment: 150,000 / 200,000 = 75.0%
Remaining to quota: $50,000

2. New vs expansion split (QTD won only)
New: $113,500 — 40,000 + 35,000 + 21,000 + 11,000 + 6,500 = 113,500 (75.7%)
Expansion: $36,500 — 20,000 (Deal-F2C7D8) + 12,000 (Deal-A8B4D6) + 4,500 (Deal-C5D9E2) = 36,500 (24.3%)
Check: 113,500 + 36,500 = 150,000

3. Active pipeline by stage (status = open, 125 deals, $1,260,390)
DS1: 20 deals, $284,621
DS2: 28 deals, $353,760
DS3: 67 deals, $552,705
DS4: 5 deals, $23,574 (Deal-1FC049 $1,920 + Deal-F9A08A $2,484 + Deal-C6D97A $3,240 + Deal-BA571A $1,080 + Deal-5408B0 $14,850 = 23,574)
DS5: 5 deals, $45,730 (Deal-403845 $9,000 + Deal-547B2B $11,200 + Deal-A2B47C $6,360 + Deal-C61CF7 $5,400 + Deal-D348E1 $13,770 = 45,730)
Check: 284,621 + 353,760 + 552,705 + 23,574 + 45,730 = 1,260,390
Note: includes 1 open deal with close_date before as-of (Deal-7A2454, DS3, $1,275, 2026-09-04). No close-date filter was given, so all status=open included.

4. Rolling 90-day DS2-to-won rate (entered_ds2 2026-06-07 to 2026-09-05)
Entered DS2 in window: 111 (8 won + 27 lost + 76 open)
Won from that cohort: 8 (Deal-A1C3E5, Deal-F2C7D8, Deal-B7D2F4, Deal-C9E1A6, Deal-A8B4D6, Deal-D4B8C2, Deal-E6F3A9, Deal-C5D9E2)
Rate (won / all entered): 8 / 111 = 7.2%
Closed-only rate (won / won+lost entered): 8 / 35 = 22.9%
Excluded from window: Deal-B3E6F1 (entered_ds2 2026-05-10). DS1 deals with blank entered_ds2 excluded from denominator — data missing for DS2 entry.

5. Wins and losses (QTD closed, close_date in quarter and <= 2026-09-05)
Wins: 8, $150,000
Losses: 27, $329,272
Loss reasons: Lost- Timing (1 year or more) 13, MIA 5, Competitor 5, Lost DM 2, Feature Request 1, Lost- Does not fit ICP (write in notes) 1
Top loss reason: Lost- Timing (1 year or more), 13/27 = 48.1%

6. Activity volume, last 30 days (sum across ae_engagements.csv, 161 rows)
Emails: 807
Calls: 112
Meetings: 128
Notes: 50
Total: 807 + 112 + 128 + 50 = 1,097

Coaching observations
1. Timing kills nearly half of losses: 13 of 27 losses are Lost- Timing (1 year or more), $40,001 high is Deal-B3ABED. Tighten DS2 qualification on timeline and defer long-horizon deals out of forecast.
2. DS2 conversion is thin: 8 wins from 111 DS2 entries in 90 days (7.2%), and only 5 deals totaling $23,574 sit in DS4. With $50,000 still to quota, prioritize advancing the $45,730 in DS5 and $23,574 in DS4 rather than adding DS1/DS2 volume.
3. Activity skews to email: 807 emails vs 112 calls and 128 meetings across 161 deal rows. Shift late-stage (DS4/DS5) touches from email to calls/meetings; 5 MIA losses suggest deals going dark on email-only follow-up.
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 $0.0196 · 66s · in 1,764 / out 3,392 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
Assumptions stated because required fields were not provided:
- Open/closed status, amount, and stage are missing from both files. Amount/stage listed as missing below, not invented.
- Active cutoff: 60 days before 2026-10-20 = 2026-08-21. Active = is_former=false AND last_engaged_date >= 2026-08-21.

Flagged deals (12 of 14 in file):

1. Deal-EC3025 / C-FDD0C7
- Amount: missing from data. Stage: missing from data.
- Active count: 1. Math: CT-047C54 2026-09-02 = active; CT-F2C1AE former = excluded.
- Personas present: champion (1)
- Personas missing: economic buyer, HR admin, IT security, finance
- Most valuable to add given stage: cannot determine - stage missing
- Unengaged fit: CT-6827DB, Chief People Officer, economic buyer

2. Deal-92D97D / C-E23238
- Amount: missing. Stage: missing.
- Active count: 1. Math: CT-01F5B4 2026-08-28 = active; CT-A902AE 2026-06-01 = >60 days, inactive.
- Present: HR admin
- Missing: economic buyer, champion, IT security, finance
- Most valuable to add: cannot determine - stage missing
- Unengaged fit: none on file

3. Deal-50D386 / C-EB10E4
- Amount: missing. Stage: missing.
- Active count: 2. Math: CT-AA41B2 2026-09-01 = active; CT-B9C35B 2026-08-25 = active. Flag reason: under-threaded (<3).
- Present: champion, HR admin
- Missing: economic buyer, IT security, finance
- Most valuable to add: cannot determine - stage missing
- Unengaged fit: CT-A1C4B3, Chief People Officer, economic buyer

4. Deal-D0D6B5 / C-32918E
- Amount: missing. Stage: missing.
- Active count: 1. Math: CT-87CED4 2026-09-02 = active; CT-DE6D7C 2026-08-19 = inactive (<2026-08-21); CT-FD70B2 2026-08-07 = inactive. Flag reason: single-threaded, also all-active in one persona.
- Present: champion
- Missing: economic buyer, HR admin, IT security, finance
- Most valuable to add: cannot determine - stage missing
- Unengaged fit: CT-1FA4DB, Chief People Officer, economic buyer

5. Deal-5BFE3B / C-535D36
- Amount: missing. Stage: missing.
- Active count: 1. Math: CT-57123B 2026-08-31 = active; CT-5CE757 2026-08-12 = inactive.
- Present: champion
- Missing: economic buyer, HR admin, IT security, finance
- Most valuable to add: cannot determine - stage missing
- Unengaged fit: none on file

6. Deal-84DBA6 / C-BE9BE0
- Amount: missing. Stage: missing.
- Active count: 2. Math: CT-BAA9D3 2026-09-02 = active; CT-60834D 2026-08-30 = active; CT-3F23A4 2026-08-20 = inactive (<2026-08-21). Flag reason: under-threaded (<3).
- Present: champion, economic buyer
- Missing: HR admin, IT security, finance
- Most valuable to add: cannot determine - stage missing
- Unengaged fit: CT-2A0169, Controller, finance

7. Deal-36C33F / C-077A0E
- Amount: missing. Stage: missing.
- Active count: 0. Math: CT-4FE556 2026-08-15 = inactive; CT-405B45 former = excluded; CT-86B22F former = excluded. Flag reason: single-threaded (0 <2).
- Present: none
- Missing: economic buyer, champion, HR admin, IT security, finance
- Most valuable to add: cannot determine - stage missing
- Unengaged fit: CT-1DB73E, Chief People Officer, economic buyer

8. Deal-885F45 / C-5E8EFB
- Amount: missing. Stage: missing.
- Active count: 1. Math: CT-51C81E 2026-08-26 = active; CT-D9A0E8 2026-08-11 = inactive.
- Present: economic buyer
- Missing: champion, HR admin, IT security, finance
- Most valuable to add: cannot determine - stage missing
- Unengaged fit: CT-B3F25D, IT Security Lead, IT security

9. Deal-FCBE5B / C-737030
- Amount: missing. Stage: missing.
- Active count: 1. Math: CT-4A5317 2026-08-29 = active.
- Present: champion
- Missing: economic buyer, HR admin, IT security, finance
- Most valuable to add: cannot determine - stage missing
- Unengaged fit: none on file

10. Deal-5408B0 / C-2AE3AA
- Amount: missing. Stage: missing.
- Active count: 1. Math: CT-D33AE4 2026-09-01 = active; CT-8742FD 2026-08-18 = inactive.
- Present: champion
- Missing: economic buyer, HR admin, IT security, finance
- Most valuable to add: cannot determine - stage missing
- Unengaged fit: CT-07FA76, Chief People Officer, economic buyer

11. Deal-C6D97A / C-5A8FC2
- Amount: missing. Stage: missing.
- Active count: 1. Math: CT-223DDC 2026-08-31 = active; CT-B03555 2026-08-20 = inactive; CT-4E8A2B 2026-08-05 = inactive. Flag reason: single-threaded, also all-active in one persona.
- Present: champion
- Missing: economic buyer, HR admin, IT security, finance
- Most valuable to add: cannot determine - stage missing
- Unengaged fit: none on file

12. Deal-F9A08A / C-0D15DF
- Amount: missing. Stage: missing.
- Active count: 1. Math: CT-931B10 2026-09-03 = active; CT-913581 2026-06-20 = inactive.
- Present: champion
- Missing: economic buyer, HR admin, IT security, finance
- Most valuable to add: cannot determine - stage missing
- Unengaged fit: CT-697541, Chief People Officer, economic buyer

Not flagged:
- Deal-4B0BEB: 3 active, 3 personas (champion, economic buyer, HR admin). CT-1543C8 2026-08-21 counted as active (>= cutoff); CT-3E135F 2026-08-09 inactive.
- Deal-D348E1: 4 active, 4 personas (champion, economic buyer, IT security, finance). CT-BEDF5E 2026-08-19 inactive.
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 $0.0385 · 66s · in 18,887 / out 3,378 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
Opener in first 5 minutes: retailer turnover story in 8 of 10 calls (8/10 = 80%). Other openers: Deal-403845 agenda-led, Deal-1E2498 pricing-led.
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."

Objection 1 - budget locked until next fiscal year: 4 of 10 calls (Deal-D348E1, Deal-547B2B, Deal-60C2C2, Deal-84DBA6). Always handled with turnover-savings funding.
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."

Objection 2 - revisit next quarter / underwater with open enrollment: 3 of 10 calls (Deal-5408B0, Deal-C61CF7, Deal-D9A12F). Always handled with 90-day one-department pilot.
Quote: "Makes sense. What if we scope a 90-day pilot with one department so you have internal data before next quarter's planning?"

Objection 3 - spreadsheet + quarterly gift cards status quo: 3 of 10 calls (Deal-403845, Deal-EDC141, Deal-1E2498). Always handled with automation + 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."

Concrete next step agreed: 7 of 10 calls (7/10 = 70%). Yes in Deal-D348E1, Deal-5408B0, Deal-547B2B, Deal-C61CF7, Deal-60C2C2, Deal-D9A12F, Deal-1E2498. No in Deal-403845, Deal-EDC141, Deal-84DBA6.
Quote: "Yes, Thursday at 2pm works — send the invite and I'll bring our HRIS manager."

Prospect-raised competitor 1 - Awardco in Deal-547B2B.
Quote: "We're also in late talks with Awardco — their rewards catalog looks bigger than yours."

Prospect-raised competitor 2 - Kudos in Deal-EDC141.
Quote: "How are you different from Kudos? Our CEO used them at her last company."

Coaching note 1: Add a fallback next step for committee / no-urgency stalls in Deal-403845, Deal-EDC141, Deal-84DBA6, where the call ended with concession and no date.

Coaching note 2: Vary the minute-14 ask when the same Thursday 2pm working session was already accepted in 7 calls, to avoid overlap and confirm the right attendees for each deal.
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 $0.1294 · 66s · in 75,470 / out 6,261 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 (close_date 2026-07-01 to 2026-09-30 inclusive)

COMMIT total: $44,729
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 total: $203,565
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

Weighted forecast (100% COMMIT + 35% BEST_CASE):
44,729 + (0.35 x 203,565) = 44,729 + 71,247.75 = $115,976.75

Counts inside quarter (54 deals):
COMMIT: 7; BEST_CASE: 24; PIPELINE: 23

Excluded (close_date outside quarter): 32 deals, $227,575 total
13,770 (1 COMMIT) + 28,240 (9 BEST_CASE) + 185,565 (22 PIPELINE) = 227,575

Top 5 BEST_CASE deals inside 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 53 of 54 in-quarter deals and why_buys_chars is 0 on 49 of 54, so qualification and accountability cannot be checked. Forecast category conflicts with stage: Deal-A5E80A is COMMIT in DS1, Deal-499BF6 is COMMIT in DS2, and Deal-6787C2 is PIPELINE in DS4. Close dates cluster with 17 of 54 on 2026-09-30, and Deal-42326B has a fractional amount of 2480.4, so unattended cutoff and summation logic would be fragile.
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 $0.1333 · 87s · in 68,062 / out 11,247 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
Activation hypothesis test (m1_users >= 5 AND m1_redemptions >= 1 vs 24-month retention, where retained = current_status = 'active'):

2x2, n = 220:

- Both signals: 47 companies, 31 active. 31/47 = 66.0% retained.
- Givers-only (m1_users >= 5, m1_redemptions = 0): 49 companies, 23 active. 23/49 = 46.9% retained.
- Redemption-only (m1_users < 5, m1_redemptions >= 1): 29 companies, 9 active. 9/29 = 31.0% retained.
- Neither: 95 companies, 38 active. 38/95 = 40.0% retained.

Check: 47 + 49 + 29 + 95 = 220. Active: 31 + 23 + 9 + 38 = 101. Matches file totals (101 active, 116 cancelled, 3 non_renewing).

Excluded from denominator: 0. No companies excluded. All 220 rows have non-missing m1_users, m1_redemptions, and current_status. The 3 non_renewing rows were counted as not retained (only 'active' = retained).

Single signal with largest retention lift: 5+ unique givers.

Arithmetic:
- Any givers signal (both + givers-only): 54/96 = 56.3% vs no givers signal: 47/124 = 37.9%. Lift = +18.3pp.
- Any redemption signal (both + redemption-only): 40/76 = 52.6% vs no redemption signal: 61/144 = 42.4%. Lift = +10.3pp.
- Within the 2x2 vs neither (40.0%): givers-only 46.9% (+6.9pp); redemption-only 31.0% (-9.0pp); both 66.0% (+26.0pp).

What this does and does not prove:
- Does show: in this 2023 cohort extract, companies with both first-month signals retained at a higher rate (66.0%) than givers-only (46.9%), neither (40.0%), or redemption-only (31.0%).
- Does not prove: causation. This is observational only, with no controls for tier, ARR, industry, integrations, or company size; no significance testing; and thresholds (5 givers, 1 redemption) were taken as given, not derived or validated here.
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 $0.0555 · 65s · in 23,160 / out 5,429 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. Billing ARR = MRR x 12.

Totals
- CRM ARR (39 company records): 603,581.76
- Billing ARR active-only (37 active subscriptions): 604,739.28
- Billing ARR all statuses (39 subscriptions): 617,897.76
- Variance, CRM minus Billing-active: 603,581.76 - 604,739.28 = -1,157.52. CRM is 1,157.52 lower.

Decomposition, sums to variance
- Status mismatch: +13,158.48
  4,905.24 + 8,253.24 = 13,158.48
- Missing records, net: -11,952.00
  CRM-only +16,497.24 minus billing-only 28,449.24 = -11,952.00
- Rounding: 0.00
  No differences <=1.00 found; all other matched accounts tie MRR x 12 to the cent.
- Other amount mismatch: -2,364.00
  -2,400.00 + 16.00 + 20.00 = -2,364.00
- Check: 13,158.48 - 11,952.00 + 0.00 - 2,364.00 = -1,157.52

Mismatched accounts
1. C-0C8323BF — CRM 4,905.24 vs billing-active 0.00, diff +4,905.24. Status: cancelled in billing, still valued in CRM. Suggested owner: CRM Admin / Account Owner to close/remove ARR.
2. C-0DC4FB8C — CRM 8,253.24 vs billing-active 0.00, diff +8,253.24. Status: cancelled in billing, still valued in CRM. Suggested owner: CRM Admin / Account Owner to close/remove ARR.
3. C-21629AA4 / SUB-0004 — CRM missing vs billing-active 28,449.24 (2,370.77 x 12), diff -28,449.24. Suggested owner: CRM Admin to create company record.
4. C-0D5BBE3A — CRM 16,497.24 vs billing missing, diff +16,497.24. Suggested owner: Billing Ops to verify subscription existence; Account Owner to validate CRM ARR.
5. C-0F7269D7 / SUB-0006 — CRM 24,396.00 vs billing 26,796.00 (2,233.00 x 12), diff -2,400.00. Suggested owner: Billing Ops + CRM Admin to correct CRM ARR.
6. C-0D66DF9E / SUB-0005 — CRM 23,200.00 vs billing 23,184.00 (1,932.00 x 12), diff +16.00. Suggested owner: CRM Admin to correct to 23,184.00.
7. C-14D70CE0 / SUB-0008 — CRM 18,200.00 vs billing 18,180.00 (1,515.00 x 12), diff +20.00. Suggested owner: CRM Admin to correct to 18,180.00.

All other matched aliases tie exactly (diff 0.00). No data missing for this reconciliation.

Term rule violations: term_months != 12 requires cf_agreement_end_date populated
- SUB-0002, C-1794A52C, term 24, active, cf_agreement_end_date blank — violation
- SUB-0019, C-22170CA1, term 36, active, cf_agreement_end_date blank — violation
Compliant non-12-month terms: SUB-000C C-0DB48281 (24, 2027-11-30), SUB-001A C-0FC4DBB8 (36, 2027-11-30).
Suggested owner for both violations: Deal Desk / Billing Ops to populate end date or correct term.
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 $0.0817 · 72s · in 37,057 / out 6,927 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: equal-weighted mean across 30 companies (sum/30). All rows are plan_tier=tier_three, so plan_tier cannot explain moves.

| KVM | 2026-08 | 2026-07 | Abs. change | Rel. change | Direction |
|---|---|---|---|---|---|
| Giving rate | 0.6027 | 0.6023 | +0.0004 | +0.07% | flat/up |
| Redemptions per user | 1.7302 | 1.7300 | +0.0002 | +0.01% | flat |
| 1:1 engagement | 0.4472 | 0.4469 | +0.0003 | +0.06% | flat |
| Pulse engagement | 0.5086 | 0.6006 | -0.0920 | -15.31% | down |

Arithmetic: giving 18.0814/30 vs 18.0689/30; redemptions 51.9049/30 vs 51.8995/30; 1:1 13.4153/30 vs 13.4066/30; pulse 15.2583/30 vs 18.0176/30.

Largest relative move: pulse engagement (-15.31%). Driven by size_band=enterprise: 0.2743 vs 0.5500, -0.2757 (-50.13%). SMB -0.22% and mid_market +0.21% are flat. All 10 enterprise aliases fell ~50%, e.g. C-0B2895EF, C-0B2213A9, C-0D6CC8E3, C-0D0B047C, C-0D3278C7, C-0FCCD2DF, C-0F6C0F34, C-8C2E8F00, C-0B827671, C-0BA71F12.
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 $0.1386 · 103s · in 48,351 / out 17,008 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
Last completed month: 2026-08. All 378 rows are dated 2026-01-01 to 2026-08-31, so YTD = through 2026-08.

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

Arithmetic:
TangoCard spend $3,230 + Tremendous $8,505 + custom $10,873 + Snappy $5,238 = $27,846

Provider mix, percent of spend (shares sum to 100):
custom: 10,873 / 27,846 = 39.05%
Tremendous: 8,505 / 27,846 = 30.54%
Snappy: 5,238 / 27,846 = 18.81%
TangoCard: 3,230 / 27,846 = 11.60%
Check: 39.05 + 30.54 + 18.81 + 11.60 = 100.00

Top 5 countries by redemptions:
US: 244
CA: 24
AU: 21
GB: 17
NL: 17
Next: SG 12; DE, FR, CH tied at 9.
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 $0.0327 · 55s · in 17,480 / out 2,437 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
8 accounts meet all 3 rules (R1 health_score <60 AND R2 churn_save_eligible_amount >0 AND R3 renewal within 120 days of 2026-09-05, i.e. on/before 2027-01-03).

Qualifying churn-save accounts:

C-0F6C0F34 - 49707.00 - renewal 2026-10-03 - health 51 - usage growing - seats_used 308/395 - champion_active false
C-0B827671 - 25365.00 - renewal 2026-11-14 - health 56 - usage declining - seats_used 113/202 - champion_active true
C-0B360C78 - 35748.00 - renewal 2026-10-28 - health 57 - usage growing - seats_used 246/327 - champion_active true
C-0B0F1BAB - 5494.00 - renewal 2026-09-23 - health 38 - usage flat - seats_used 238/363 - champion_active false
C-0CA21961 - 16829.00 - renewal 2026-12-28 - health 58 - usage flat - seats_used 84/325 - champion_active true
C-0E9C27D1 - 41235.00 - renewal 2026-09-24 - health 39 - usage flat - seats_used 134/157 - champion_active true
C-0CEF69FD - 32621.00 - renewal 2026-11-21 - health 53 - usage growing - seats_used 97/136 - champion_active false
C-0D3278C7 - 17602.00 - renewal 2026-11-12 - health 54 - usage declining - seats_used 126/380 - champion_active true

Total at stake:
49707 + 25365 + 35748 + 5494 + 16829 + 41235 + 32621 + 17602 = 224601.00

Plays (usage revival, executive touch, commercial concession):
Play-assignment criteria were not provided. eligibility_rules.csv contains only R1/R2/R3. No playbook mapping usage_trend_3m / champion_active / seat utilization to a play was included, so a play cannot be assigned under documented rules without inventing criteria. Relevant signals per qualifying account are listed above.

At-risk (R1 true) but NOT qualifying:

C-0BC71BDD - health 55 - fails R2: churn_save_eligible_amount 0.00
C-0BA71F12 - health 52, amount 6824.00 - fails R3: renewal 2027-04-11 after 2027-01-03
C-0F6694C3 - health 43 - fails R2: 0.00
C-0BE96399 - health 54 - fails R2: 0.00
C-0F876796 - health 47, amount 19958.00 - fails R3: renewal 2027-02-06 after 2027-01-03
C-0FCCD2DF - health 43 - fails R2: 0.00
C-10A56B0F - health 54 - fails R2: 0.00
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 $0.0289 · 40s · in 20,092 / out 885 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 for C-0DDFC9A7

Seat coverage (licensed / headcount):
150 / 400 = 0.375 = 37.5%

Usage health:
- Monthly active users rose every month: 88 (2026-03) -> 95 -> 102 -> 110 -> 118 -> 126 (2026-08).
- August utilization of licensed seats: 126 / 150 = 0.84 = 84%.

Headroom at current per-seat rate:
- Per-seat rate: 9,000.00 / 150 = $60.00
- Seat headroom: 400 - 150 = 250 seats
- ARR headroom: 250 x $60.00 = $15,000.00

Who replied / can they buy:
- Maria S., People Operations Coordinator, last engaged 2026-09-02.
- Cannot buy: states "I'm not the right person for purchasing decisions here."

Right buyer from contacts:
- Dana R., VP People, last engaged 2026-05-18. Maria confirms "Budget and seat expansion sit with Dana R."

Reply email (98 words):

Hi Maria,

Thank you — glad the team loves Bonusly and that the feed is busy weekly.

Your active users climbed from 88 in March to 126 in August, so Dana asking about usage numbers makes sense.

Would you be open to making that introduction to Dana R.? Happy to share a brief usage snapshot for her, no pressure on timing or budget.

If easier, I can send a short forwardable note for you.

Thanks again for pointing me the right 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 $0.0324 · 60s · in 19,984 / out 1,735 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
C-0D284E42 mid-onboarding brief — signup 2026-08-11

Complete (with data field):
- Slack integrated: integration_slack = 2026-08-12 (1 day after signup: 12 - 11 = 1)
- Allowance set: allowance_set = 2026-08-13 (2 days after signup: 13 - 11 = 2)
- Admins added: admins_added = 2 — target number not provided, so sufficiency unknown
- First recognition given: first_recognition_at = 2026-08-15 14:22 (4 days after signup: 15 - 11 = 4)

Not complete (field blank):
- HRIS integration: integration_hris = [blank]
- First redemption: first_redemption_at = [blank] — 20 days without redemption as of last usage date 2026-09-04 (31-15=16 + 4 = 20)

Early engagement signals — active_givers, 2026-08-11 to 2026-09-04:
- Start 3 → end 15. Change: 15 - 3 = 12. Growth: 15 / 3 = 5.0x
- 25 days total. Sum = 214. Average = 214 / 25 = 8.56
- First 7 days (08-11 to 08-17): 3+3+4+4+5+4+7 = 30. Avg = 30 / 7 = 4.29
- Last 7 days (08-29 to 09-04): 11+13+11+13+13+15+15 = 91. Avg = 91 / 7 = 13.0
- Signal: consistent upward trend, no reversal to starting level. Peak 15 reached 09-03 and held 09-04.
- Missing data: total seats / invited users not provided, so adoption % cannot be calculated. Recognition count, redemption catalog status, HRIS blocker reason not provided.

Three things to cover on the call:
1. HRIS: integration_hris blank while Slack connected 2026-08-12 — confirm owner, blocker, and date.
2. First redemption: first_redemption_at blank 20 days after first recognition 2026-08-15 — demo redeem flow and set a redemption goal.
3. Lock in giver habit: active_givers grew 3 to 15, avg 4.29 to 13.0 — confirm allowance fit (set 2026-08-13) and admin coverage (2 added) to sustain it.
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 $0.0551 · 73s · in 24,527 / out 5,746 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 — trusted window: 2026-09-15 to 2026-11-24 (20 renewals from files)

Date trust rule: is_multi_year=true → trust chargebee_terms.csv. ChurnZero is known wrong for multi-year.
Result: trust Chargebee for all 5 multi-year accounts. 15 annual accounts agree.

DISAGREEMENTS FLAGGED — 5 total:
C-0B7D2C30: CZ 2026-09-10 vs CB 2026-09-15. Use 2026-09-15. Trust CB, 36-mo multi-year.
C-0BCDB8C2: CZ 2027-09-18 vs CB 2026-09-18. Use 2026-09-18. Trust CB, 36-mo multi-year.
C-0D2AB865: CZ 2026-09-10 vs CB 2026-09-22. Use 2026-09-22. Trust CB, 24-mo multi-year.
C-0BBE3E60: CZ 2027-09-26 vs CB 2026-09-26. Use 2026-09-26. Trust CB, 24-mo multi-year.
C-0F5D2323: CZ 2026-09-10 vs CB 2026-09-29. Use 2026-09-29. Trust CB, 24-mo multi-year.

All others: CZ = CB, no disagreement. Date used = agreed date.

Seat utilization = seats_used / seats x 100
3-mo trend = (2026-08 active – 2026-06 active) / 2026-06 active x 100

RENEWALS:

C-0B7D2C30 | Dana Mercer | $65,901 | Use 2026-09-15 | Util 274/476=57.6% | Trend 84-97=-13, -13.4% declining | HIGH — 57.6% seats and -13.4% 3-mo decline with 12-mo fall 155 to 84.
C-0BCDB8C2 | Cole Ingram | $54,427 | Use 2026-09-18 | Util 232/424=54.7% | Trend 110-127=-17, -13.4% declining | HIGH — 54.7% seats and -13.4% 3-mo decline with 12-mo fall 200 to 110.
C-0D2AB865 | Elena Sinclair | $38,022 | Use 2026-09-22 | Util 250/407=61.4% | Trend 109-125=-16, -12.8% declining | HIGH — 61.4% seats and -12.8% 3-mo decline with 12-mo fall 199 to 109.
C-0BBE3E60 | Dana Mercer | $30,993 | Use 2026-09-26 | Util 74/114=64.9% | Trend 33-39=-6, -15.4% declining | HIGH — 64.9% seats and -15.4% 3-mo decline with 12-mo fall 63 to 33.
C-0F5D2323 | Cole Ingram | $90,647 | Use 2026-09-29 | Util 111/390=28.5% | Trend 18-20=-2, -10.0% flat-low | HIGH — 28.5% seats, lowest utilization cohort with flat 18-21 active users.
C-0EC6999D | Elena Sinclair | $79,419 | Use 2026-10-03 | Util 31/112=27.7% | Trend 15-17=-2, -11.8% flat-low | HIGH — 27.7% seats, lowest utilization cohort with flat 14-17 active users.
C-0B20DB64 | Dana Mercer | $21,770 | Use 2026-10-07 | Util 214/378=56.6% | Trend 294-294=0, 0.0% flat | MEDIUM — 56.6% seats but stable ~294-298 active users for 12 mo.
C-0BBC4E7A | Cole Ingram | $56,374 | Use 2026-10-10 | Util 228/337=67.7% | Trend 139-142=-3, -2.1% flat | MEDIUM — 67.7% seats with flat 139-142 active users.
C-0FD551AB | Elena Sinclair | $48,815 | Use 2026-10-14 | Util 210/376=55.9% | Trend 126-123=+3, +2.4% flat | MEDIUM — 55.9% seats with flat 122-127 active users.
C-0F9F8F13 | Dana Mercer | $46,230 | Use 2026-10-18 | Util 199/352=56.5% | Trend 182-185=-3, -1.6% flat | MEDIUM — 56.5% seats with flat 181-185 active users.
C-0BC34584 | Cole Ingram | $16,740 | Use 2026-10-22 | Util 327/494=66.2% | Trend 106-104=+2, +1.9% flat | MEDIUM — 66.2% seats with flat 101-106 active users.
C-0B7A7546 | Elena Sinclair | $35,062 | Use 2026-10-25 | Util 182/205=88.8% | Trend 63-64=-1, -1.6% flat | LOW — 88.8% seats with stable 58-65 active users.
C-0B369871 | Dana Mercer | $85,128 | Use 2026-10-29 | Util 317/422=75.1% | Trend 333-326=+7, +2.1% growing | LOW — 75.1% seats and +2.1% 3-mo growth with 12-mo rise 289 to 333.
C-0B144C78 | Cole Ingram | $30,899 | Use 2026-11-02 | Util 169/224=75.4% | Trend 106-101=+5, +5.0% growing | LOW — 75.4% seats and +5.0% 3-mo growth.
C-0FC4DBB8 | Elena Sinclair | $94,732 | Use 2026-11-05 | Util 356/464=76.7% | Trend 193-189=+4, +2.1% growing | LOW — 76.7% seats and +2.1% 3-mo growth with 12-mo rise 168 to 193.
C-0D5BBE3A | Dana Mercer | $39,740 | Use 2026-11-09 | Util 85/102=83.3% | Trend 91-88=+3, +3.4% growing | LOW — 83.3% seats and +3.4% 3-mo growth.
C-0FB9D5AF | Cole Ingram | $63,158 | Use 2026-11-13 | Util 144/199=72.4% | Trend 176-173=+3, +1.7% growing | LOW — 72.4% seats and +1.7% 3-mo growth.
C-0B344485 | Elena Sinclair | $64,384 | Use 2026-11-16 | Util 224/287=78.0% | Trend 244-238=+6, +2.5% growing | LOW — 78.0% seats and +2.5% 3-mo growth with 12-mo rise 211 to 244.
C-0CB2C1B4 | Dana Mercer | $40,628 | Use 2026-11-20 | Util 386/473=81.6% | Trend 49-47=+2, +4.3% flat | LOW — 81.6% seats with stable 43-50 active users.
C-22170CA1 | Cole Ingram | $45,646 | Use 2026-11-24 | Util 251/294=85.4% | Trend 146-143=+3, +2.1% flat-up | LOW — 85.4% seats with stable-to-growing 130-148 active users.

TOTALS:
Total ARR renewing = 65,901+54,427+38,022+30,993+90,647+79,419+21,770+56,374+48,815+46,230+16,740+35,062+85,128+30,899+94,732+39,740+63,158+64,384+40,628+45,646 = $1,048,715

ARR at risk (HIGH-rated) = 65,901+54,427+38,022+30,993+90,647+79,419 = $359,409
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 $0.0239 · 64s · in 3,739 / out 3,836 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
Total tickets: 80 (IC-460001 to IC-460080).

Ranked by ARR exposure. ARR affected = sum of distinct account ARRs.

1. HRIS provisioning failure - new hires not created
Count: 12 / 80 = 15.0%
Distinct accounts: 3
ARR affected: 36,000 + 48,000 + 30,000 = $114,000
Accounts: C-0B2213A9, C-0DDFC9A7, C-0F6C0F34
Example tickets: IC-460059, IC-460060
Recommendation: Fix silent HRIS sync skip and add provisioning failure alerting.

2. Rewards redemption / checkout failure - points deducted, no gift card
Count: 18 / 80 = 22.5%
Distinct accounts: 7
ARR affected: 8,900 + 10,700 + 9,600 + 8,700 + 9,600 + 10,300 + 11,000 = $68,800
Accounts: C-0CEF69FD, C-0B827671, C-0FCCD2DF, C-0F876796, C-0D9CA315, C-0B0F1BAB, C-14264ABD
Example tickets: IC-460025, IC-460024
Recommendation: Stop deducting points on failed checkout and fix fulfillment email/code delivery.

3. SINGLE-ACCOUNT NOISE: Recurring billing / seat-count / renewal-tier errors
Count: 16 / 80 = 20.0%
Distinct accounts: 1
ARR affected: $52,000
Account: C-0E9C27D1
Example tickets: IC-460071, IC-460069
Recommendation: Audit and correct C-0E9C27D1 seat count and renewal tier immediately; do not treat as broad billing bug.

Broad patterns continue:

4. Points / recognition not posting to balances
Count: 20 / 80 = 25.0%
Distinct accounts: 9
ARR affected: 3,500 + 2,700 + 2,500 + 2,900 + 3,400 + 4,500 + 4,500 + 4,200 + 2,900 = $31,100
Accounts: C-0D3278C7, C-0BE96399, C-0DD0626C, C-0B2895EF, C-0D284E42, C-0BF20542, C-0D0B047C, C-0D6CC8E3, C-21FEBCBB
Example tickets: IC-460004, IC-460016
Recommendation: Investigate points posting delay/failure after delivery.

5. Slack integration - sync, re-auth, slash command failures
Count: 14 / 80 = 17.5%
Distinct accounts: 4
ARR affected: 4,400 + 3,900 + 5,400 + 5,200 = $18,900
Accounts: C-0B843542, C-0BA71F12, C-10A56B0F, C-8C2E8F00
Example tickets: IC-460041, IC-460046
Recommendation: Fix Slack sync toggle reset, re-auth persistence, and slash command errors.

Check: 12 + 18 + 16 + 20 + 14 = 80.
Tags ignored: billing/how-to/feedback tags cut across all text themes and do not match content.
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 $0.0096 · 49s · in 665 / out 1,373 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
Ranked social proof for prospect C-82AF3719 (Technology, Mid-Market, employee_recognition, NA-West):

Scoring: 1 point per exact match on industry, size_band, use_case, region.

Eligible pool = has_case_study=true only (8 companies).

1. C-64171065 – Score: 3/4
1 (industry: Technology) + 1 (size_band: Mid-Market) + 1 (use_case: employee_recognition) + 0 (region: NA-East vs NA-West) = 3
Drove match: industry, size_band, use_case

2. C-A13C193D – Score: 3/4
1 (industry: Technology) + 1 (size_band: Mid-Market) + 0 (use_case: retention vs employee_recognition) + 1 (region: NA-West) = 3
Drove match: industry, size_band, region

3. C-11C31562 – Score: 3/4
0 (industry: Manufacturing vs Technology) + 1 (size_band: Mid-Market) + 1 (use_case: employee_recognition) + 1 (region: NA-West) = 3
Drove match: size_band, use_case, region

Tie-break: all three score 3/4; ranked by industry match first, then use_case match.
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 $0.0639 · 72s · in 29,375 / out 5,687 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 to 2026-08 inclusive. All 122 contact rows fall in this range.

PAID CHANNELS

paid_search
spend = 6,000 x 6 = 36,000
SQMs = 40
SQOs = 18
cost per SQM = 36,000 / 40 = 900
cost per SQO = 36,000 / 18 = 2,000
SQM-to-SQO rate = 18 / 40 = 45.0%
pipeline amount = 18 x 40,000 = 720,000
pipeline per dollar = 720,000 / 36,000 = 20.00

linkedin_ads
spend = 4,000 x 6 = 24,000
SQMs = 25
SQOs = 8
cost per SQM = 24,000 / 25 = 960
cost per SQO = 24,000 / 8 = 3,000
SQM-to-SQO rate = 8 / 25 = 32.0%
pipeline amount = 8 x 12,000 = 96,000
pipeline per dollar = 96,000 / 24,000 = 4.00

webinars
spend = 1,500 x 6 = 9,000
SQMs = 12
SQOs = 5
cost per SQM = 9,000 / 12 = 750
cost per SQO = 9,000 / 5 = 1,800
SQM-to-SQO rate = 5 / 12 = 41.7%
pipeline amount = 5 x 12,000 = 60,000
pipeline per dollar = 60,000 / 9,000 = 6.67

paid_social
spend = 3,000 x 6 = 18,000
SQMs = 0
SQOs = 0
cost per SQM = undefined (spend with zero SQMs, not zero)
cost per SQO = undefined (zero SQOs from zero SQMs)
SQM-to-SQO rate = undefined (0 / 0)
pipeline amount = 0
pipeline per dollar = 0 / 18,000 = 0.00

ORGANIC / ZERO-SPEND CHANNELS
No spend rows provided for these channels. Reporting volume as SQM count.

organic_search
volume = 30
SQOs = 10
SQO rate = 10 / 30 = 33.3%
pipeline = 10 x 9,000 = 90,000

referral
volume = 15
SQOs = 6
SQO rate = 6 / 15 = 40.0%
pipeline = 6 x 8,000 = 48,000

DATA QUALITY FLAGS - SQO date precedes SQM date:
CT-000044, linkedin_ads, SQM 2026-07-23, SQO 2026-07-18
CT-000041, linkedin_ads, SQM 2026-06-14, SQO 2026-06-09
Both affect linkedin_ads SQO count of 8; if excluded, linkedin_ads would be 6 SQOs, not 8.

REALLOCATION RECOMMENDATION
Reallocate paid_social spend (18,000 with zero SQMs, undefined efficiency) to paid_search first, webinars second. paid_search has highest pipeline per dollar at 20.00 vs 6.67 webinars vs 4.00 linkedin_ads, and highest SQM-to-SQO rate at 45.0% vs 41.7% webinars vs 32.0% linkedin_ads. Reduce linkedin_ads pending fix of the 2 flagged date inversions.

Confidence: moderate-low. Sample sizes over 6 months are small: SQOs = 18 paid_search, 8 linkedin_ads, 5 webinars, 10 organic_search. A 1-2 deal swing changes webinars and linkedin_ads rates materially, and 2/8 linkedin_ads SQOs are date-flagged.
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 $0.0165 · 52s · in 1,313 / out 2,792 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
# Battlecard: Rivally (updated)

## One-line positioning
Points-based recognition feed [S02], with fit noted in mid-market [S04].

## Pricing (newer source wins, conflict noted)
- Current list: $7 per user/month, annual billing required, Recognition Starter tier [S17, 2026-08-12 pricing_page].
- Conflict: older list was $5 per user/month, annual billing required [S03, 2026-01-20 pricing_page], still shown as $5 on 2026-04-01 [S08, pricing_page]. Newer $7 source wins.
- Deal-quoted checks: $6.50/user/mo to a 500-seat prospect, annual term [S13, 2026-06-02 call_notes]; $7/user/mo list, offered 15% discount for 3-year term [S18, 2026-08-14 call_notes].
- Add-on pricing: Rivally Pulse is priced as add-on, not bundled [S23, 2026-09-01 press].
- Old card claim $5/user/mo as of 2026-01 is superseded by [S17].

## Where they win
- Fast setup: setup took under a week [S04].
- Slack integration worked out of the box [S04].
- EU enterprise fit: pitched EU data residency [S05]; EU data residency generally available with Dublin office opening [S15]; strong for distributed EU teams, multi-language support praised [S12].
- Engaging recognition feed [S02] [S16].
- Support response time praised, under 4 hours [S22].
- Old card claim Strong in EU enterprise with multi-language support is re-sourced by [S12] [S15].

## Where we win
- Analytics depth: limited analytics [S02]; reporting dashboards basic compared to enterprise tools [S07]; analytics exports are CSV-only, made migration hard [S20].
- Head-to-head proof on analytics: 800-seat prospect picked Bonusly over Rivally citing analytics depth [S25].
- Identity/admin: lacks SCIM provisioning, manual user management is painful [S10]; admin tooling lags peers [S16]; admin console still lacks bulk recognition editing [S24].
- Rewards breadth in region: rewards catalog in EMEA is thinner than US catalog [S14].

## Objections and responses
- Objection: Prospect evaluating Rivally and Bonusly, Rivally pitched EU data residency [S05].
  Response: Acknowledge EU data residency is generally available with Dublin office [S15] and multi-language strength for distributed EU teams [S12]; counter with analytics depth [S02] [S07] [S25], SCIM gap [S10], and EMEA catalog thinness [S14].
- Objection: Rivally quoted $6.50/user/mo [S13] or $7 list with 15% discount for 3-year term [S18].
  Response: Anchor to current $7 list with annual billing required [S17]; note Pulse is add-on cost, not bundled [S23].
- Excluded as non-factual rep opinion, not usable as competitor fact: AE opinion that UI is clunky [S09]; AE opinion that Rivally is discounting aggressively [S21].

## Recent changes
- Series C $40M led by Northgate Ventures [S01, 2025-11-04 press].
- Launches Rivally Pulse lightweight engagement survey add-on [S06, 2026-03-05 press]; exits beta, priced as add-on not bundled [S23, 2026-09-01 press].
- Hires ex-Workday VP EMEA to lead European expansion [S11, 2026-05-09 press].
- Opens Dublin office, announces EU data residency generally available [S15, 2026-07-01 press].
- Pricing page updated to $7 per user/month [S17, 2026-08-12 pricing_page].
- Microsoft Teams app v2 in public preview [S19, 2026-08-20 press].

## Our 12-month win/loss record vs Rivally
Source: deals_with_competitor.csv, 2025-09 through 2026-08 inclusive, 20 deals.

Arithmetic: 13 wins + 7 losses = 20 total. 13/20 = 65.0% win rate.

- 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
- By month: 2025-09: 1-1; 2025-10: 2-0; 2025-11: 1-1; 2025-12: 1-1; 2026-01: 2-0; 2026-02: 2-0; 2026-03: 1-1; 2026-04: 0-2; 2026-05: 0-1; 2026-06: 1-0; 2026-07: 1-0; 2026-08: 1-0. Sum = 13-7.

## Old card carryover verification
- Points-based recognition for mid-market: re-sourced [S02] [S04].
- Lacks a Slack integration: UNVERIFIED / contradicted - only re-source found says Slack integration worked out of the box [S04].
- Acquired by WorkHuman in 2025: UNVERIFIED - no snippet re-sources this.
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 $0.0535 · 70s · in 25,240 / out 3,765 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
Per sequence totals [sent = sum steps]:

New Logo Nurture: sent 1,386. Open 490/1,386=35.4%. Reply 90/1,386=6.5%. Meetings 27/1,386=1.9%. Weakest: Step 3 - 28.0% open, 4.2% reply, 1.4% meeting.

Expansion Nurture: sent 875. Open 565/875=64.6%. Reply 59/875=6.7%. Meetings 12/875=1.4%. Weakest: Step 3 - 34.5% open, 4.4% reply, 1.1% meeting.

Cold Outbound - HR Leaders: sent 1,785. Open 545/1,785=30.5%. Reply 8/1,785=0.45%. Meetings 0/1,785=0%. Weakest: Step 3 - 22.0% open, 0.17% reply, 0 meetings all steps.

Cold Outbound - People Ops: sent 1,163. Open 340/1,163=29.2%. Reply 29/1,163=2.5%. Meetings 6/1,163=0.5%. Weakest: Step 3 - 21.2% open, 1.6% reply, 0.3% meeting.

Tracking error: Expansion Nurture Step 2 opened 340 > sent 300 =113.3% open. Impossible; open logging or sent suppression broken. Inflates total open rate.

Audience overlap: contact_keys appear in two sequences. New Logo Nurture + Expansion Nurture: e.g. CT-000301, CT-000624. Cold Outbound - HR Leaders + Cold Outbound - People Ops: e.g. CT-000849, CT-000884, CT-001033, CT-001101, CT-001103, CT-001153, CT-001159, CT-001217, CT-001227, CT-001255, CT-001277, CT-001285, CT-001345. Causes double-counting and competing sends.

Under 2% reply: only Cold Outbound - HR Leaders at 0.45%. Failure mode: open holds then reply collapses 0.83% to 0.17% with 0 meetings. Offer/message mismatch, not deliverability. No content data to isolate further.

Fix first Cold Outbound - HR Leaders: pause Steps 2-3, rewrite Step 1 offer and rebuild list excluding People Ops overlap.
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 $0.0099 · 51s · in 273 / out 1,551 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 (66 of 92 days elapsed = 71.7%)

Elapsed: 66/92 = 0.7174 = 71.7% expected attainment for linear pace.

1. SQMs: QTD 230 / Target 300 / Delta 230-300 = -70 / Attainment 230/300 = 76.7% / Prorated expected 300*66/92 = 215.2 / +14.8 vs prorated / Pace: ahead

2. SQOs: QTD 84 / Target 120 / Delta 84-120 = -36 / Attainment 84/120 = 70.0% / Prorated expected 120*66/92 = 86.1 / -2.1 vs prorated / Pace: behind

3. DS2s: QTD 40 / Target 75 / Delta 40-75 = -35 / Attainment 40/75 = 53.3% / Prorated expected 75*66/92 = 53.8 / -13.8 vs prorated / Pace: behind

4. Closed-lost MIA rate: QTD 5/25 = 0.20 = 20.0% / Target 0.10 = 10.0% / Delta 0.20-0.10 = +0.10 (+10.0 pts, lower_better) / Pace: behind (above target)

5. Same-quarter closes: QTD 10 / Target 20 / Delta 10-20 = -10 / Attainment 10/20 = 50.0% / Prorated expected 20*66/92 = 14.3 / -4.3 vs prorated / Pace: behind

6. Active pipeline coverage: QTD $3,000,000 / Target $4,000,000 / Delta -$1,000,000 / Coverage 3,000,000/4,000,000 = 75.0% / Prorated expected 4,000,000*66/92 = $2,869,565 / +$130,435 vs prorated / Pace: ahead

What moved this week: No week-over-week data was provided — only QTD actuals — so weekly movement cannot be determined from the data provided.
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 $0.0395 · 78s · in 20,137 / out 3,360 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
Do not report Q3 as 115,976.75 without re-qualification. Forecast equals 44,729 from 7 COMMIT plus 0.35 times 203,565 from 24 BEST_CASE which is 71,247.75 for 115,976.75 across 54 of 86 deals. All 7 COMMIT deals lack why-buys and owner is blank on 85 of 86, so require documented reasons before committing.
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 $0.0314 · 50s · in 20,079 / out 1,492 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: July 28 demo recap + pricing for 150 seats

Hi,

Following up on my Aug 5 recap of the July 28 demo with pricing for 150 seats — wanted to make sure you saw it.

You'd flagged strong interest in automated milestone awards and the points catalog to replace manual tracking of birthdays and anniversaries.

Are you open to a brief reconnect this week to confirm if this is still a priority?

Thanks,
Alex Franklin

Deal-0D2F7A — Body: 65 words < 120
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 $0.1001 · 69s · in 68,723 / out 1,931 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 closed the week with 46 SQMs against a 52 target (52-46=6 short; 46/52=88.5% to goal), with heart for the progress and focus for what's next. The webinar channel led the way with 18 of those SQMs (18/46=39.1%; 46-18=28 from all other channels), showing how shared learning turns into real momentum. Cost per SQM on paid search held at 150 dollars, keeping efficient growth front and center.

Sales kept deals moving with 14 SQOs converted and 9 DS2 meetings set (14+9=23 buyer advances), turning great conversations into clear next steps. New pipeline created was 310000 dollars (310000/14=22142.86 dollars per SQO), adding meaningful fuel for the quarter ahead. Same-quarter close count for the week was 3, a welcome win worth celebrating together.

CS protected customer love with 2 renewals saved, helping teams stay connected to recognition that matters. Team NPS moved to 61, a strong signal that everyday support and partnership are landing. There are 3 open red-flag accounts heading into next week, where proactive care can make all the difference.

PLG welcomed 412 new signups with activation at 31 percent (412x0.31=127.7, about 128 activated; 412-128=284 not yet activated), inviting more teams to experience belonging from day one. 38 companies reached the aha moment of 10 recognition gives (38/412=9.2% of signups), the point where giving becomes habit and culture starts to stick.
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 $0.0282 · 55s · in 20,012 / out 741 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: 5 deals, $550,000
Arithmetic: 180,000 + 95,000 + 140,000 + 60,000 + 75,000 = 550,000

Active:

- Apex Rewards Co — 2 sourced opps, both in DS1. Deal-DDAAF2 ($180,000) + Deal-2CF33E ($95,000) = $275,000. Co-webinar locked for 09-15. UTM Source = Partner confirmed.
- HRCloud Partners — 1 sourced opp. Deal-F1CDA5 ($140,000). Opp moved to DS2; integration referral closed the security review.
- CultureBridge — 2 sourced opps, both early stage. Deal-096E1D ($60,000) + Deal-067213 ($75,000) = $135,000. Sourced via lunch-and-learn.
- WorkWell Group — 0 sourced deals this period. Joint playbook restart planned for Q4; planning call booked for 09-09.

Quiet:

- Recogniq — no sourced deals this period.
- KudosWave — no sourced deals this period.
- PeopleFirst Advisors — no sourced deals this period.
- TotalPerk — no sourced deals this period.
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 $0.0077 · 54s · in 431 / out 997 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 and SCIM user provisioning are 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 provided excerpts. No excerpt in [P1]-[P8] addresses ISO 27001 certification. Would require compliance certification documentation.

Q9: Unanswerable from the provided excerpts. No excerpt in [P1]-[P8] states a contractual uptime SLA percentage. Would require MSA / SLA terms.

Q10: Unanswerable from the provided excerpts. No excerpt in [P1]-[P8] addresses HIPAA Business Associate Agreements. Would require legal / HIPAA compliance documentation.
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 $0.1315 · 81s · in 83,966 / out 6,237 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
1. OVERLAPPING ALWAYS-TRIGGERS
Severity: CRITICAL | Action: MERGE
Evidence:
- comms-drafter description claims: "write me an email," "draft a follow-up," "help me reply," "what should I say," "bump email," "contract nudge"
- email-drafter description claims: "write me an email," "draft a follow-up," "help me reply to this," "what should I say," "bump email," "contract nudge"
Overlap = 6 identical trigger phrases ("write me an email" / "draft a follow-up" / "help me reply" / "what should I say" / "bump email" / "contract nudge"). Both claim full external-copy scope.
Proposal: MERGE into one skill; keep the other as alias. Do not keep two ALWAYS triggers for email copy.

2. CIRCULAR DELEGATION CHAIN
Severity: WARNING | Action: REVIEW
Evidence:
- email-drafter Lane marker: "point them to the deal-strategy-coach skill"
- deal-strategy-coach Manager email section: "use the `email-drafter` skill which automatically retrieves your Gmail signature"
Chain: email-drafter -> deal-strategy-coach -> email-drafter
Proposal: REVIEW and set one-way precedence (strategy may call drafter, drafter must not loop back; replace drafter-to-coach pointer with a non-invoking suggest line).

3. DANGLING DELEGATION TARGET
Severity: WARNING | Action: REVIEW
Evidence: these invoked targets have no manifest row and no FILE block:
- `bonusly-brand` (invoked by comms-drafter Step 0, email-drafter, sales-forecast STEP 3, signalforge-claim-compressor)
- `prospect-research-multithreading` (invoked by comms-drafter, email-drafter, deal-strategy-coach Cross-skill handoff)
- `signalforge-reports` (invoked by pipeline-intelligence-report Phase 5, weekly-pipeline-report Step 4)
- `skill-orchestrator` (invoked by analysis-validator Sec 11, signalforge-feedback Activation Checklist)
- `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` (all invoked by analysis-validator Sec 12.4)
Arithmetic: 14 manifest names exist; 0 of the above targets are in that 14.
Proposal: REVIEW: either add missing skill files + manifest rows, or UPDATE_BODY to remove the delegation.

4. VERSION CONFLICT
Severity: WARNING | Action: UPDATE_BODY
Evidence: analysis-validator.SKILL.md contains:
- Header: "**Version:** 3.6" / "**Last Updated:** May 9, 2026 (v3.6"
- Sec 7 trail template: "Validator:      analysis-validator v3.2"
- Footer: "analysis-validator v3.6 · May 9, 2026"
Conflict: v3.6 vs v3.2 in same file.
Proposal: UPDATE_BODY to v3.6 everywhere; v3.6 survives (header + footer + Changelog 3.6 entry agree; v3.2 is single stale trail line).

5. MANIFEST DESCRIPTIONS >1,024 CHARS
Severity: INFO | Action: REVIEW
Arithmetic: description_chars = [656, 897, 996, 792, 965, 676, 945, 1004, 1006, 962, 1006, 708, 762, 656]. Count >1024 = 0. Max = 1006 (pipeline-intelligence-report, signalforge-claim-compressor). 0/14 exceed.
Proposal: REVIEW: no TRIM_DESC needed; limit respected.

6. HARDCODED IDS / DATES / PERSON NAMES IN BODIES
Severity: WARNING | Action: UPDATE_BODY
Evidence (samples, quoted exactly):
- analysis-validator Sec 12.3: person names "Bryce Harmon", "Hugo Lindqvist", "Dana Mercer", "Alex Franklin", "Cole Ingram", "Gavin Porter", "Alaina Loori", "Shealagh Coughlin", "Amani Phipps", "John Thomas", "Yasmin Wahid" + owner IDs "83155923", "1520255671" + stage IDs "150582536", "150582537" + dates "May 4, 2026", "May 9, 2026"
- deal-strategy-coach: hardcoded page "https://bonusly1612893911.atlassian.net/wiki/spaces/SignalForg/pages/2257879045/AE+Excellence+Playbook+April+2026"
- partner-digest: Cloud ID `73fe98de-a4a3-4869-9f8a-bb1eeed4cf7f`, Space ID `1958248479`, folder ID `2286616609`, pages `2286321666`, `2265382925`, `2236940297`, Slack `from:<@U03QLMBL7AR>`, dates "May 16, 2026", "May 19, 2026", "June 2, 2026"
- sales-forecast: Space ID `2232811524`, Parent `2232582148`, spreadsheets `1CLZeOsElVDF_LF0ZG_t2nfwvhnZ6bpwqM_nX3WEYzcw`, `1ENuaEcCuLjdKhMvp8FK3Ys1ek5Aw9ZuOZhsHJJFoB_k`, name "Alaina", date "July 9, 2026"
- signalforge-feedback: Page ID `2295136266`, Parent `2234417154`, Build Log `2247295002`
- stale-pipeline-report: channel ID `C0561C1JCPJ`, owner `55483190`, URL `https://app.hubspot.com/contacts/1973303/deal/{id}`
- pipeline-intelligence-report / next-to-close: org `1973303`, stage IDs `150582536, 150582537, 150582538, 150582539, 1175632767`
Proposal: UPDATE_BODY to move IDs/dates/roster to live lookups or a reference file; rewrite nothing now.

7. MANIFEST DRIFT BOTH DIRECTIONS
Severity: INFO | Action: REVIEW
Arithmetic: manifest rows = 14. FILE blocks provided = 14. Intersection by file/name = 14.
Files with no manifest row = 14-14 = 0.
Manifest rows with no file = 14-14 = 0.
Proposal: REVIEW: no DELETE_SKILL or manifest add needed; set in sync.
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 $0.0311 · 61s · in 20,176 / out 1,385 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-2026-08-29-reward-queue

1. [M01] Acknowledge and take IC
- Who: Bryce Harmon
- Action: PagerDuty alert for reward-worker queue depth > 10k acknowledged; taking IC.
- Verification: Not stated in thread.
- State change: No.

2. [M02] Check queue depth
- Who: Farid Osman
- Command: `bundle exec rake sidekiq:queue_depth`
- Result: 48,213 pending jobs; normal is under 500.
- Verification: Command output itself.
- State change: No.

3. [M03] Check dead set — NEEDS CONFIRMATION (method/command not stated)
- Who: Farid Osman
- Action: Observed dead set has 112 jobs, all Redis::TimeoutError from around 13:58.
- Verification: Not stated.
- State change: No.

4. [M04] Pause enqueue
- Who: Farid Osman
- Command: `bin/rails runner 'FeatureFlag.disable(:auto_recognition_enqueue)'`
- Verification: Not stated in thread.
- Rollback (stated in M04): `bin/rails runner 'FeatureFlag.enable(:auto_recognition_enqueue)'`

5. [M05] Clear dead set — NEEDS CONFIRMATION (exact command not stated)
- Who: Elena Sinclair
- Action: "cleared out the dead set" while in the console.
- Verification: Not stated.
- Rollback: None stated in thread — NEEDS CONFIRMATION.

6. [M06] Scale workers up 3 → 6
- Who: Bryce Harmon
- Command: `kubectl scale deployment/reward-worker --replicas=6` (was 3)
- Verification: Not stated at this step; see M07/M08 for later queue drop.
- Rollback (stated in M06): `kubectl scale deployment/reward-worker --replicas=3`

7. [M07] Monitor queue drain
- Who: Farid Osman
- Observation: Queue depth down to 9,400 and falling ~1,200/min.
- Verification: Queue depth reading; method/command not stated — NEEDS CONFIRMATION if same as `bundle exec rake sidekiq:queue_depth`.
- State change: No.
- Arithmetic: 48,213 − 9,400 = 38,813 jobs drained.

8. [M08] Verify drain to zero
- Who: Cole Ingram
- Commands/Checks: `bundle exec rake sidekiq:queue_depth` now returns 0; error rate in Datadog back to baseline.
- Verification: Command output + Datadog baseline.
- State change: No.

9. [M09] Re-enable enqueue
- Who: Bryce Harmon
- Command: `bin/rails runner 'FeatureFlag.enable(:auto_recognition_enqueue)'`
- Verification: 40 new jobs processed cleanly in the next 3 minutes. Verification method not stated — NEEDS CONFIRMATION.
- Rollback: None separately stated; reverse is M04 disable command, but thread does not label it as rollback for this step.

10. [M10] Scale workers back down 6 → 3, resolve
- Who: Bryce Harmon
- Command: `kubectl scale deployment/reward-worker --replicas=3`
- Verification: Queue stable at 0. Verification method not stated — NEEDS CONFIRMATION if same queue-depth command.
- Rollback: None stated in thread — NEEDS CONFIRMATION.
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 $0.0335 · 68s · in 21,320 / out 1,601 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

Cascade in order:
1. 14:01:12Z reward-service Redis timeout to redis-primary:6379
2. 14:01:20Z, 14:01:30Z, 14:01:40Z reward-service: retry exhausted for RewardGiveJob (3x)
3. 14:01:40Z sidekiq: RewardGiveJob failed: Redis::TimeoutError; retrying in 60s
4. 14:02:28Z sidekiq: RewardGiveJob failed; retrying
5. 14:02:30Z sidekiq WARN: Queue reward depth above 10,000
6. 14:03:05Z api-gateway ERROR: 502 upstream timeout calling reward-service /gives
7. 14:03:30Z web-app ERROR: Give form submission failed: upstream 502 from api-gateway
8. 14:03:31Z to 14:06:52Z repeated sidekiq RewardGiveJob failures + api-gateway 502s + web-app Give form failures
9. 14:22:10Z reward-service INFO: Redis connection restored; resuming job processing
10. 14:24:45Z sidekiq INFO: Queue reward depth below 500

Arithmetic:
Error count in slice = 4 reward-service + 6 sidekiq + 5 api-gateway + 4 web-app = 19 ERROR rows
sidekiq_jobs.csv = 12 RewardGiveJob + 4 RecognitionDigestJob = 16 failed jobs
Outage window first error to restore = 14:01:12Z to 14:22:10Z = 20m 58s
Queue drop = above 10,000 to below 500 = >9,500 cleared

Service and job involved:
Service: reward-service (downstream impact: sidekiq, api-gateway, web-app)
Job: RewardGiveJob

Datadog query to confirm first error:
service:reward-service level:ERROR "Redis::TimeoutError" from 2026-09-03T14:00:00Z to 2026-09-03T14:02:00Z

Logs do not show:
- Root cause of redis-primary:6379 timeout; no Redis server/infra metrics, deploys, or network events
- No deal/company aliases provided in either file
- No customer/user impact counts, no Give values, no resolution action taken
- sidekiq_jobs.csv lists 4 RecognitionDigestJob Redis failures (14:02:36Z-14:05:50Z) with no matching rows in datadog_logs.csv slice
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 $0.0094 · 58s · in 466 / out 1,388 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
Listed flags (9 total: 5 on, 4 off):

1. recognition_streaks_v2 — on — segment:beta_companies — 42 companies
   Controls: if enabled, StreakTracker.record(give) in app/models/recognition.rb

2. points_budget_guardrails — on — all_companies — 220 companies
   Controls: if enabled, BudgetService.new(company).enforce!(giver, points) in app/services/budget_service.rb

3. slack_dm_nudges — on — segment:region_na — 87 companies
   Controls: return unless enabled, then SlackDm.send_nudge(user) in app/jobs/nudge_job.rb

4. redeem_flow_redesign — off — targeted_list — 12 companies
   Controls: if enabled render RedeemV2Component.new(company), else render RedeemV1Component.new(company) in app/controllers/redeem_controller.rb

5. analytics_dashboard_v3 — on — segment:tier_three — 65 companies
   Controls: if enabled, @dashboard = AnalyticsV3.new(company) in app/controllers/analytics_controller.rb

6. ms_teams_app_v2 — off — targeted_list — 9 companies
   Controls: if enabled, TeamsAppV2.install(company) in app/services/teams_installer.rb

7. legacy_give_modal — off — segment:legacy_plan — 14 companies
   Controls: no code reference in provided excerpt

8. survey_boosters_q3 — on — segment:legacy_plan — 7 companies
   Controls: no code reference in provided excerpt

9. paused_offboard_cleanup — off — [no targeting_rules] — 0 companies
   Controls: no code reference in provided excerpt
   No targets: targeting_rules empty, company_count 0

Arithmetic (company_count sums exactly as exported):
On: 42 + 220 + 87 + 65 + 7 = 421
Off: 12 + 9 + 14 + 0 = 35
Total: 421 + 35 = 456
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 $0.0096 · 59s · in 519 / out 1,407 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 — [PARTY A] — GREEN — Reasoning: standard mutual 2-year term with 3-year survival (2+3=5 years total confidentiality window), full carve-outs and Delaware law with "No license, no obligation to proceed, no exclusivity" — driven by "Term: two (2) years from the Effective Date; obligations survive for three (3) years after termination."

NDA-2 — [PARTY B] — YELLOW — Reasoning: embedded non-solicit extending 18 months past 3-year term (3 years + 18 months = 4.5 years total restriction window) requires counsel review — driven by "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 — [PARTY C] — RED — Reasoning: one-way 5-year term with 3-year non-compete, Ireland exclusive jurisdiction, and missing carve-outs requires full legal review — driven by "For a period of three (3) years following the Effective Date, Recipient shall not, directly or indirectly, engage in or provide services to any business that competes with Discloser's business." plus "No confidentiality carve-outs are stated in this Agreement." and "This Agreement shall be governed by and construed in accordance with the laws of the Republic of Ireland, and the parties submit to the exclusive jurisdiction of its courts."