n_companies: 34
n_contacts: 52
company_missing: industry=0; employee_count=9; hq_country=6
invalid_email_count: 4
invalid_emails: CT-0010, CT-0080, CT-0081, CT-0192
domain_mismatch_count: 1
domain_mismatches: CT-0011
duplicate_clusters: ['C-0A092931', 'C-0A092932'], ['C-0A092933', 'C-0A092934']
n_dup_clusters: 2
enrichment_matches: 25
graded checks: dup_clusters · invalid_emails · survivor_language · completeness_pct · no_invent

Model's actual answer

show the model's full answer

CRM AUDIT — completeness + duplicate + enrichment review

DATA PROVIDED
- companies.csv: 34 rows
- contacts.csv: 52 rows
- zoominfo_enrichment.csv: 25 rows
- deals.csv: NOT PROVIDED

============================================================
CRITICAL GAP
============================================================
No deals.csv in the inputs. Cannot audit owner, stage, amount, close date, or why-buys. Cannot compute true pipeline $ at stake. All fix-priority rankings below use employee_count + contact coverage as a proxy.

============================================================
COMPANY COMPLETENESS (n=34)
============================================================
industry       : 34/34 = 100.0% raw fill; 18/34 = 52.9% need casing/spacing normalization
                 "tech" x 3   (C-66D1FC, C-44EA29, C-60C75F)
                 "Tech " x 4  (C-425E2A, C-BA969B, C-93C8BF, C-C9BB20) -- trailing space
                 "Technology" x 9
                 "health care" x 2 (lowercase) (C-7BBDFA, C-50D386)
employee_count : 25/34 = 73.5%; empty for C-EC3025, C-96039F, C-44EA29, C-D04904,
                 C-B23205, C-60C75F, C-7BBDFA, C-50D386, C-93C8BF
hq_country     : 28/34 = 82.4%; empty for C-2D1F1B, C-D73B89, C-44EA29, C-D04904,
                 C-2C60E5, C-EE9FFB
                 Filled values split across 3 spellings: US x 9, USA x 6, United States x 2
                 Canada x 8, UK x 3

============================================================
CONTACT COMPLETENESS (n=52)
============================================================
email  : 48/52 = 92.3% valid; 4 invalid (end with bare "@", no domain)
title  : 40/52 = 76.9%
persona: 37/52 = 71.2%

5 contacts have BOTH title and persona empty: CT-0000, CT-0022, CT-0092, CT-0132, CT-0162

14 of 34 companies have ZERO contacts:
  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

============================================================
DUPLICATE COMPANY CLUSTERS (shared domain)
============================================================
Cluster 1 -- acme-corp.com
  C-0A092931 (Technology, 500, US)
  C-0A092932 (tech, 510, USA)
  Survivor: C-0A092931 -- canonical industry spelling, canonical HQ spelling.
  Action: move any C-0A092932 contacts (none in this extract) to C-0A092931;
          archive C-0A092932; resolve employee_count drift (500 vs 510) with owner.

Cluster 2 -- globex.io
  C-0A092933 (SaaS, 200, US)
  C-0A092934 (Technology, 200, US)
  Survivor: ambiguous -- no enrichment row to disambiguate. If "SaaS" matches your
  ICP, keep C-0A092933; otherwise keep C-0A092934 (industry spelling matches the
  1500-emp "Technology" pattern). Either way, archive the other.

No other shared-domain clusters. No name-variant clusters (aliases are opaque IDs).

============================================================
INVALID EMAILS (4) and DOMAIN MISMATCHES (1)
============================================================
INVALID -- email ends in "@" with no domain:
  CT-0010  C-66D1FC      user0@          VP People, champion    900 emp
  CT-0080  C-92D97D      user0@          HR admin               50 emp
  CT-0081  C-92D97D      user1@          (no title, no persona) 50 emp
  CT-0192  C-425E2A      user2@          People Ops, champion   50 emp

DOMAIN MISMATCH -- email domain != company domain:
  CT-0011  C-66D1FC      user1@other-domain.com   VP People, champion   900 emp
           (company domain 66d1fc.com; contact "domain" col already says 66d1fc.com)

============================================================
ENRICHMENT MATCHES -- FIELLS AVAILABLE
============================================================
25 ZI rows. 8 employee_count gaps fill from ZI:

  domain       CRM emp   ZI emp   company
  ec3025.com   empty     400      C-EC3025
  96039f.com   empty     400      C-96039F
  44ea29.com   empty     400      C-44EA29
  d04904.com   empty     400      C-D04904
  b23205.com   empty     400      C-B23205
  60c75f.com   empty     400      C-60C75F
  7bbdfa.com   empty     400      C-7BBDFA
  50d386.com   empty     400      C-50D386

0 HQ fills from ZI: every row with a non-empty ZI HQ matches the CRM value after
US/USA/United States normalization; the 4 CRM-empty HQ rows (2d1f1b.com, d73b89.com,
44ea29.com, d04904.com) are also empty in ZI.

0 industry fills from ZI: CRM has a value for all 25 domains that ZI covers.

============================================================
ENRICHMENT DISAGREEMENTS (both sources have a value)
============================================================
Industry -- 10 rows disagree (CRM tech/Technology/Tech vs ZI "Computer Software"):

  domain       CRM             ZI                  company
  66d1fc.com   tech            Computer Software   C-66D1FC
  ec3025.com   Technology      Computer Software   C-EC3025
  44ea29.com   tech            Computer Software   C-44EA29
  92d97d.com   Technology      Computer Software   C-92D97D
  d04904.com   Technology      Computer Software   C-D04904
  77a95a.com   Technology      Computer Software   C-77A95A   <- 1500 emp
  aa8dda.com   Technology      Computer Software   C-AA8DDA   <- 1500 emp
  b25f40.com   Technology      Computer Software   C-B25F40
  60c75f.com   tech            Computer Software   C-60C75F
  425e2a.com   Tech (space)    Computer Software   C-425E2A

Recommendation: adopt ZI "Computer Software" as canonical for these 10 (more
specific, matches enrichment). Do NOT auto-overwrite C-77A95A and C-AA8DDA
without owner sign-off -- both are 1500-emp and a category swap can re-segment
a large book. Note C-0A092933 (SaaS) has no ZI row to validate against.

Employee count -- 0 disagreements among the 16 rows where both have a value (all match).

HQ country -- all 17 disagreements are spelling-only (US/USA/United States); normalize
to a single value.

============================================================
UNFILLABLE GAPS (no ZI row)
============================================================
C-93C8BF  93c8bf.com  employee_count empty, HQ Canada        -- no ZI row
C-EE9FFB  ee9ffb.com  HQ empty, 1500 emp Healthcare           -- no ZI row
C-BA969B  ba969b.com  no ZI row, industry "Tech " unverified  -- normalization only
C-332637  332637.com  no ZI row, industry "Technology"       -- normalization only
C-C9BB20  c9bb20.com  no ZI row, industry "Tech " unverified  -- normalization only
C-2C60E5  2c60e5.com  HQ empty -- ZI HQ also empty           -- no source
C-2D1F1B  2d1f1b.com  HQ empty -- ZI HQ also empty           -- no source
C-D73B89  d73b89.com  HQ empty -- ZI HQ also empty           -- no source

These cannot be auto-filled from the provided enrichment. Flag for manual research
or a second enrichment source.

============================================================
TOP 10 FIXES BY IMPACT (proxy = employee_count + contact coverage)
============================================================
No deals.csv, so $ at stake is approximate. Ranked by exposure to large companies
and active contact counts.

 1. Merge acme-corp.com cluster -- C-0A092931 + C-0A092932.
    ~1010 emp combined, currently double-counted in any pipeline rollup by company.
    Pick C-0A092931 as survivor; archive C-0A092932.

 2. Merge globex.io cluster -- C-0A092933 + C-0A092934.
    400 emp combined, same double-count issue. Pick a survivor (C-0A092934 unless
    "SaaS" is your ICP category) and archive the other.

 3. Bulk-normalize hq_country to one canonical value across the 17 affected rows
    (US x 9, USA x 6, United States x 2). Required for any geo-segmented pipeline
    report. Touches 5 of the 7 1500-emp companies: C-77A95A, C-E51FB7, C-D0662E,
    C-0A092931, plus C-0A092934 (200).

 4. Fix C-66D1FC contact block (900 emp, all 3 contacts broken):
    CT-0010 invalid email, CT-0011 email-domain mismatch to other-domain.com,
    CT-0012 ok but re-validate. Also normalize industry "tech" -> canonical.

 5. Fill employee_count from ZI for the 8 companies above (sorted by contact count):
    C-96039F (3 contacts, Finance), C-B23205 (3, Healthcare), C-D04904 (3,
    Technology, also missing HQ), C-44EA29 (3, also missing HQ + industry
    disagreement), C-EC3025 (2, also industry disagreement), C-60C75F (2, also
    industry disagreement), C-7BBDFA (0), C-50D386 (0).

 6. Normalize industry for the 10 ZI-disagreement rows to "Computer Software",
    pending owner sign-off on C-77A95A and C-AA8DDA (both 1500 emp).

 7. Backfill contacts for 14 zero-contact companies. Priority by size:
    1500-emp: C-EE9FFB (Healthcare, also missing HQ), C-B97B4E (Retail, UK),
              C-332637 (Technology, UK)
    900-emp : C-2D7423 (Manufacturing, USA), C-C9BB20 (Tech, UK)

 8. Fix remaining invalid emails CT-0080, CT-0081 (C-92D97D, 50 emp, 2/3 contacts
    broken) and CT-0192 (C-425E2A, 50 emp). Small accounts but blocks all
    outbound.

 9. Resolve unfillable gaps that have no ZI row: C-93C8BF employee_count,
    C-EE9FFB HQ country (1500 emp). Manual research or a second enrichment source
    required -- do not invent.

10. Obtain deals.csv and re-run this audit with owner/stage/amount/close date/
    why-buys completeness and true pipeline-$ prioritization. Until then, fix
    ordering above is proxy-based.

============================================================
ALSO-RECOMMENDED (outside top 10 by strict size ranking)
============================================================
- Normalize "tech" / "Tech " / "Technology" spelling across the 16 rows that have
  no ZI disagreement (C-BA969B, C-332637, C-2D1F1B, C-31ED2A, C-92D97D already in
  disagreements, etc.) -- pick one canonical label.
- Bulk-fill missing title + persona on the 5 fully-empty contacts (CT-0000,
  CT-0022, CT-0092, CT-0132, CT-0162) and the 9 with persona-only empty.
- 14 zero-contact companies (item 7) extend beyond the 5 listed; remaining are
  C-2C60E5 (340 emp Healthcare), C-7BBDFA, C-50D386, C-BA969B (50 emp), and the
  4 duplicate-cluster aliases (C-0A092931, C-0A092932, C-0A092933, C-0A092934)
  which will resolve via merges 1 and 2.

Classification per transcript

TG-001 / Deal-EC3025
- Prospect (IT Security Lead): "We need SCIM user provisioning — our security team will not approve any tool that requires manual account management."
- Docs: "SCIM user provisioning... NOT currently listed as supported capabilities" (no plan carries it).
- Verdict: REAL GAP. Capability absent from the entire product.
- Amount: not present in transcripts_gaps.csv.

TG-002 / Deal-D0D6B5
- Prospect (HRIS Manager): "Our payroll runs on ADP Workforce Now and I don't see ADP anywhere in your integration list — that's a dealbreaker for us."
- Docs: "ADP Workforce Now integrations are NOT currently listed as supported capabilities" (HRIS list is Workday, BambooHR, Gusto, Rippling on Pro+).
- Verdict: REAL GAP. ADP integration absent.
- Amount: not present in transcripts_gaps.csv.

TG-003 / Deal-CFE7F4
- Prospect (VP People): "I tried to build a custom report combining recognition frequency with turnover and couldn't — is that possible on our current plan?"
- Docs: Custom report builder = Enterprise only. Capability exists, gated by tier.
- Verdict: PLAN GATE. Not a product gap; an upsell conversation.
- Amount: not present in transcripts_gaps.csv.

TG-004 / Deal-84DBA6
- Prospect (People Ops Manager): "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."
- Docs: Slack integration is available on all plans. Failure is adoption/training, not capability.
- Verdict: ROLLOUT / ENABLEMENT issue. Not a product gap.
- Amount: not present in transcripts_gaps.csv.

TG-005 / Deal-36C33F
- Mobile-app reference is from Alex Franklin (rep). Prospect (HR Manager) replied: "Good to know. The web version should be fine for our office staff for now."
- Rep language does not count as prospect voice; prospect accepted the workaround.
- Verdict: excluded — not raised by the prospect.
- Amount: not present in transcripts_gaps.csv.

Summary — real product gaps (prospect-raised only)

1. SCIM user provisioning — Deal-EC3025. Prospect quote: "We need SCIM user provisioning — our security team will not approve any tool that requires manual account management." Not on any tier per product docs. Amount: missing from source data.
2. ADP Workforce Now integration — Deal-D0D6B5. Prospect 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." Not in supported HRIS list. Amount: missing from source data.

show the model's full answer

FIRST FIVE MINUTES — LEAD
8/10 calls open with the same case-study line on minute 0 (TT-001 Deal-D348E1, TT-002 Deal-5408B0, TT-003 Deal-547B2B, TT-005 Deal-C61CF7, TT-006 Deal-60C2C2, TT-007 Deal-EDC141, TT-008 Deal-D9A12F, TT-010 Deal-84DBA6): "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."

The other 2 calls opened differently:
- TT-004 Deal-403845 (minute 0): "I put together a short agenda — security review first, then pricing."
- TT-009 Deal-1E2498 (minute 0): "You asked for straight pricing last time, so let's start there."

THREE MOST COMMON OBJECTIONS (with count out of 10)
1. Budget locked — 4/10 (TT-001, TT-003, TT-006, TT-010)
   Prospect line: "Honestly, budget is locked until next fiscal year — I can't add a new line item right now."
   Alex's response (verbatim, all 4 times): "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."

2. Revisit next quarter — 3/10 (TT-002, TT-005, TT-008)
   Prospect line: "This is good, but can we revisit it next quarter? Open enrollment starts in October and we're underwater."
   Alex's response (verbatim, all 3 times): "Makes sense. What if we scope a 90-day pilot with one department so you have internal data before next quarter's planning?"

3. Spreadsheet status quo — 3/10 (TT-004, TT-007, TT-009)
   Prospect line: "We already do recognition with a spreadsheet and quarterly gift cards — why would we change?"
   Alex's response (verbatim, all 3 times): "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."

NEXT STEP RATE
7 of 10 calls locked a concrete next step = 70%.
Next-step-yes: TT-001 Deal-D348E1, TT-002 Deal-5408B0, TT-003 Deal-547B2B, TT-005 Deal-C61CF7, TT-006 Deal-60C2C2, TT-008 Deal-D9A12F, TT-009 Deal-1E2498 — all closed with "Yes, Thursday at 2pm works — send the invite and I'll bring our HRIS manager."
Next-step-no: TT-004 Deal-403845 ("Understood — I'll leave it with you."), TT-007 Deal-EDC141 ("Fair enough."), TT-010 Deal-84DBA6 ("Understood, thanks for the candor.").

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

Note: Workhuman appeared once, but raised by Alex himself in TT-005 Deal-C61CF7 (minute 2), not by a prospect — excluding per the brief.

COACHING NOTES
1. All 3 lost next steps share one trigger: a committee or "no urgency" objection. In TT-004 Deal-403845 ("We need to see what the budget committee says…"), TT-007 Deal-EDC141 ("I need to think about it — there's no urgency on our side."), and TT-010 Deal-84DBA6 ("We'll have to wait for the committee…"), Alex folded with "Understood" or "Fair enough" and offered no fallback. Coach him to always leave with a soft next step — e.g., "Let's lock 20 minutes on [specific date] once the committee lands" — so the 30% of calls that drift still produce a re-entry point.
2. The minute-0 opener is identical on 8/10 calls, and on the 3 lost deals (TT-004, TT-007, TT-010) it didn't pre-empt the objection that actually killed the next step. The two calls with custom openers (security-first on Deal-403845, pricing-first on Deal-1E2498) both still hit the spreadsheet objection but Deal-1E2498 converted — suggests a tailored opener per deal stage beats the one-size-fits-all case study.

REDEMPTION SECTION — Weekly Rewards Report through 2026-08
Last completed month: 2026-08 (data spans 2026-01 through 2026-08; 378 rows total)

REDMEMPTIONS, SPEND & REACHERS
  Redemptions:           378
  Total spend:           $27,846.00
  Unique redeemers:      236
  Redemptions/redeemer:  1.60  (378 / 236)

PROVIDER MIX (% of spend)
  custom       $10,873.00   39.05%
  Tremendous    $8,505.00   30.54%
  Snappy        $5,238.00   18.81%
  TangoCard     $3,230.00   11.60%
  TOTAL        $27,846.00  100.00%

TOP 5 COUNTRIES BY REDEMPTIONS
  US  244
  CA   24
  AU   21
  GB   17
  NL   17   (tie with GB at 17)

Notes:
- Scope is all rows with redeemed_at on or before 2026-08-31; no rows fall outside that window in the file.
- US = 64.6% of redemptions (244 / 378); remaining 36.4% spread across 11 other countries (CA, AU, GB, NL, SG, DE, FR, CH, IN, BR, NZ).
- "custom" is the single largest provider by spend (39.05%) despite only ~10% of redemption count, because the average ticket ($324 avg on custom rows) is far higher than TangoCard/Tremendous/Snappy.

⚠ Stream stalled mid tool-call (execute_code); the action was not executed. Ask me to retry if you want to continue.THEMES (ranked by ARR exposure, broad patterns only):

1. BILLING / INVOICE ERRORS — 14 tickets, 26.9% — 2 accounts (C-0E9C27D1 ×13, C-0B0F1BAB ×1) — $733,000 ARR — IC-460071, IC-460066 — C-0E9C27D1 alone accounts for 13/14 (92.9%) of all billing tickets in the quarter, with $676,000 ARR; root cause likely a single misconfigured seat-count mapping or stale renewal tier.

2. HRIS PROVISIONING FAILURES — 11 tickets, 21.2% — 4 accounts (C-0B2213A9 ×5, C-0DDFC9A7 ×3, C-0F6C0F34 ×2, C-0E9C27D1 ×1) — $444,000 ARR — IC-460059, IC-460060 — High-severity cross-account pattern (not single-account noise): provisioning log shows no errors yet users don't get created, suggesting a silent integration break.

3. REDEMPTION / CHECKOUT FAILURES — 13 tickets, 25.0% — 6 accounts (C-0B827671 ×3, C-0FCCD2DF ×2, C-0F876796 ×3, C-14264ABD ×3, C-0CEF69FD ×2, C-0B0F1BAB ×1) — $80,200 ARR — IC-460025, IC-460022 — Spread across 6 accounts, mid-tier revenue impact; recommendation: instrument the checkout success rate by step.

4. SLACK INTEGRATION BREAKAGE — 11 tickets, 21.2% — 4 accounts (C-0BA71F12 ×5, C-10A56B0F ×3, C-0B843542 ×2, C-8C2E8F00 ×1) — $30,300 ARR — IC-460041, IC-460046 — Disconnection + re-auth issues + slash command errors cluster around C-0BA71F12; possible Slack-side OAuth regression.

5. POINTS NOT POSTING — 12 tickets, 23.1% — 8 accounts (C-0D3278C7 ×3, C-0BF20542 ×2, C-0D0B047C ×2, C-0D6CC8E3 ×2, C-0BE96399 ×1, C-0DD0626C ×1, C-0B2895EF ×1, C-0D284E42 ×1) — $42,200 ARR — IC-460016, IC-460014 — Broadest account spread (8), low ARR per ticket; weekend/job-timing pattern suggests a stuck batch or scheduler failure.

SINGLE-ACCOUNT NOISE (excluded from ranked themes; reported separately):
- None — all five themes represent multi-account patterns.

NOTES:
- Total tickets analyzed: 52.
- Total distinct accounts: 20.
- Total ARR at risk across all 52 tickets: $863,800 (sum of unique ticket-level ARR values).
- Rankings exclude single-account issues by design; no such issues exist in this dataset (every theme has ≥4 accounts except Billing, which is concentrated in C-0E9C27D1).
- Billing's apparent concentration in one account is itself a finding, not a disqualifier — the pattern recurs at one account across multiple weeks, indicating a systemic misconfiguration on that account, not random noise.

WEEKLY MARKETING GOALS UPDATE — Q3-2026
Days elapsed: 66 of 92 (71.74% of quarter); days remaining: 26

SQMs
  QTD actual:   230
  Target:       300
  Delta:        -70  (76.67% of target)
  Expected pace at day 66: 300 × 0.71739 = 215.22
  Pace:         AHEAD  (+14.78 vs expected; +6.87%)

SQOs
  QTD actual:   84
  Target:       120
  Delta:        -36  (70.00% of target)
  Expected pace at day 66: 120 × 0.71739 = 86.09
  Pace:         ON PACE  (-2.09 vs expected; -2.43%, within ±5%)

DS2s
  QTD actual:   40
  Target:       75
  Delta:        -35  (53.33% of target)
  Expected pace at day 66: 75 × 0.71739 = 53.80
  Pace:         BEHIND  (-13.80 vs expected; -25.66%)

Closed-lost MIA rate
  QTD actual:   5 / 25 = 0.20  (20.00% of closed-lost deals)
  Target:       0.10  (10.00%, lower_better)
  Delta:        +0.10  (10.00 percentage points over target; 2.0× the target rate)
  Expected pace at day 66: 0.10 × 0.71739 = 0.07174 (7.17%)
  Pace:         BEHIND  (actual 20.00% vs expected 7.17%; +178.8% worse than pace)

Same-quarter closes
  QTD actual:   10
  Target:       20
  Delta:        -10  (50.00% of target)
  Expected pace at day 66: 20 × 0.71739 = 14.35
  Pace:         BEHIND  (-4.35 vs expected; -30.31%)

Active pipeline coverage
  QTD actual:   $3,000,000
  Target:       $4,000,000
  Delta:        -$1,000,000  (75.00% of target)
  Expected pace at day 66: $4,000,000 × 0.71739 = $2,869,565
  Coverage:     0.75× target  ($3.00M held against $4.00M needed)
  Pace:         ON PACE  (+$130,435 vs expected; +4.55%, within ±5%)

Pace classification rule: ±5% of linear expected = on pace; outside band = ahead/behind per direction.

What moved this week: The files provided (marketing_qtd.csv, targets.csv, quarter_meta.csv) contain only QTD-cumulative figures — there is no week-over-week or daily series, no deal-level deltas, no channel/source breakdown, and no date stamp within the week. I therefore cannot identify what moved this week from the data supplied. Flagging this explicitly as missing rather than inferring: to produce a weekly movement paragraph, inputs would need either a previous-week snapshot for delta computation or a row-level event feed (new SQMs/SQOs/DS2s created, deals closed-won or closed-lost, MIA flags set) with timestamps.

Marketing closed the week of August 31 at 46 SQMs against a target of 52 — about 88% of plan. The webinar channel delivered 18 of those SQMs, roughly 39% of weekly volume. Paid search cost per SQM held steady at $150, a number the team has defended for several weeks running. Net read: a small miss on volume, but with the highest-cost channel stable and webinars doing the top-funnel work.

Sales turned those marketing SQMs into real downstream motion: 14 SQOs booked, 9 DS2 meetings set (a 64% SQO-to-DS2 conversion), and $310,000 in new pipeline created. Same-quarter close count landed at 3 for the week. The SQO-to-meeting ratio is the standout — most weeks it's the leakier part of the funnel, and this one is tightening. Pipeline dollars per SQO run about $22.1K ($310,000 ÷ 14), a healthy floor heading into September.

Customer Success saved 2 renewals this week — quiet wins, but renewals saved compound. Team NPS moved up to 61, comfortably above the 50 line and the strongest print we've shown in the digest recently. Three accounts are flagged red heading into next week, so Monday is already mapped. CS is doing the steady work that keeps the retention curve honest.

PLG brought in 412 new signups with a 31% activation rate — roughly 128 activated accounts (412 × 0.31) — and 38 of those companies crossed the aha-moment threshold of 10 recognition gifts given. That 38 is the number to watch: it's the smallest cohort that has actually felt the product working. Signup volume is healthy; the activation gap is the lever to pull in next week's experiments.

partner-digest-sectionPASS 1.00

? The Question

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 $0.0035 · 121s · in 451 / out 2,040 tok

✓ all_partners✓ pipeline_amount✓ quiet_one_line

✓ Correct Answer (ground truth)

roster: partner_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)
per_partner_pipeline: {"Apex Rewards Co": {"count": 2, "amount": 275000}, "HRCloud Partners": {"count": 1, "amount": 140000}, "CultureBridge": {"count": 2, "amount": 135000}}
partner_sourced_pipeline: count=5; amount=550000
quiet_partners: Recogniq, KudosWave, PeopleFirst Advisors, TotalPerk
active_no_deals: WorkWell Group
rule: every roster partner named exactly once; quiet partners get one line; no partners outside roster.
graded checks: all_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 (UTM Source = Partner)
  Deals:   5
  Amount:  $550,000
  Breakdown by partner:
    Apex Rewards Co       2 deals   $275,000
    HRCloud Partners      1 deal    $140,000
    CultureBridge         2 deals   $135,000
    WorkWell Group        0 deals   $0

ACTIVE PLAYS

Apex Rewards Co — active
  Pipeline: 2 deals, $275,000
    Deal-DDAAF2  $180,000
    Deal-2CF33E  $95,000
  Status: Co-webinar with Apex locked for 09-15. Both sourced opps logged this
  period via UTM Source = Partner; per Slack, both in DS1.

HRCloud Partners — active
  Pipeline: 1 deal, $140,000
    Deal-F1CDA5  $140,000
  Status: Integration referral closed the security review; per Slack, the opp
  moved to DS2. One sourced opp this period.

CultureBridge — active
  Pipeline: 2 deals, $135,000
    Deal-096E1D  $60,000
    Deal-067213  $75,000
  Status: Lunch-and-learn produced two sourced opps; per Slack, both early
  stage.

WorkWell Group — active
  Pipeline: 0 deals, $0
  Status: No sourced deals this period. Joint playbook restart targeted for
  Q4; planning call booked for 09-09.

QUIET PARTNERS

Recogniq         — no Slack activity this period; no sourced pipeline.
KudosWave        — no Slack activity this period; no sourced pipeline.
PeopleFirst Advisors — no Slack activity this period; no sourced pipeline.
TotalPerk        — no Slack activity this period; no sourced pipeline.

Notes
  - Pipeline count and dollars reconcile to partner_deals.csv: 5 rows where
    utm_source = Partner, summing to $550,000.
  - Deal stage references (DS1, DS2, "early stage") are taken from Slack and
    are not present as fields in partner_deals.csv.

rfp-answer-from-policyPASS 1.00

? The Question

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 $0.0023 · 15s · in 446 / out 1,012 tok

✓ answered_count✓ unanswerable_flagged✓ citations✓ no_guess

✓ Correct Answer (ground truth)

answered_count: 7
answer_map: {"Q1": ["P1"], "Q2": ["P1"], "Q3": ["P2"], "Q4": ["P3"], "Q5": ["P4"], "Q6": ["P5"], "Q7": ["P6"]}
unanswerable: Q8, Q9, Q10
valid_citation_ids: P1, P2, P3, P4, P5, P6, P7, P8
rule: unanswerable items must be flagged and the document that would answer them named; do not guess.
graded checks: answered_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. A copy 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. The current subprocessor list is published at trust.bonusly.com. Customers receive 30 days notice before a new subprocessor is added. [P4]

Q6. Yes. SAML 2.0 single sign-on is supported on the Enterprise plan (SCIM user provisioning is also available on that plan). [P5]

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

Q8. Unanswerable from the provided excerpts. The policy excerpts do not reference an ISO 27001 certification. The document that would resolve this is Bonusly's certifications/compliance page on the trust portal (trust.bonusly.com) or a current ISO 27001 certificate of registration shared under NDA.

Q9. Unanswerable from the provided excerpts. No uptime SLA percentage is stated in the policy excerpts. The document that would resolve this is the executed Master Service Agreement or Service Level Agreement addendum between Bonusly and your organization.

show the model's full answer

NDA-1 — GREEN
Driving clauses: §3 (full five-part carve-out set: public availability, prior knowledge, third-party receipt, independent development, legal compulsion) and §4 (Delaware governing law).
Reasoning: Mutual, 2-year term + 3-year survival is standard, no embedded restrictive covenants, Delaware is the routine US corporate forum — clean standard approval.

NDA-2 — YELLOW
Driving clause: §4 "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."
Reasoning: Embedded non-solicit (18-month tail, contact-nexus scope) restricts hiring of anyone met during the evaluation; otherwise the NDA is standard (mutual, Delaware, full carve-outs). Send to counsel for scope carve-outs (general advertising, unsolicited applications, employees initiated contact) and to confirm enforceability under Delaware law before signing.

NDA-3 — RED
Driving clauses: §2 "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"; §3 "the laws of the Republic of Ireland, and the parties submit to the exclusive jurisdiction of its courts"; §4 "No confidentiality carve-outs are stated in this Agreement."
Reasoning: Three stacked problems — (1) 3-year blanket non-compete with no product/market/geography limitation is overbroad and likely unenforceable but creates litigation exposure; (2) Irish governing law + exclusive Irish jurisdiction is unusual for a US SaaS counterparty and adds dispute cost/complexity; (3) zero carve-outs means any incidentally received information could be claimed as confidential, choking normal product development and competitive hiring. Full legal review required; do not sign as-is.