Run ID: 20260629-080558 | Monitoring: 104 | Graduated (ref): 7
- Pricing experiment (abc-pro-plan-pricing-v3) is structurally blocked: control arm deprecated 58 days ago with only 46 exposures. Catastrophe threshold cannot be evaluated. Needs product decision: reframe, re-enable control, or ship best performer.
- Blog migration working: impressions +76%, clicks +62% since launch. Daily avg now ~3,900 impressions. Target of 100 is a smoke placeholder (actual: 153K cumulative).
- Internal linking SEO: city pages lost 9.2% clicks while hospital pages gained 56.8%. The registered KPI was never instrumented in 93 days. Needs human input.
- 4 launches ready for graduation: map-markers-grey-blocks (75d, zero regressions), email-pii-removal (94d, codebase audit clean), mobile-address-input-styling (79d, iOS share +6.7pp), map-z-index-ordering (47d, all 10 tests green).
- Refund copy change (day 53) shows >= 37% refund rate drop with no conversion damage. Continue monitoring.
- @gsingal — Decide pricing experiment future: reframe/re-enable control/ship (abc-pro-plan-pricing-v3)
- @gsingal — Investigate city page ranking regression (internal-linking-seo, -9.2% clicks)
- @gsingal — Instrument or retire internal-linking KPI (93 days, never measured)
- @gsingal — Graduate 4 launches: map-markers-grey-blocks, email-pii-removal, mobile-address-input-styling, map-z-index-ordering
- @gsingal — Recalibrate blog-migration and blog-content-sitemap KPI targets (smoke values)
- @gsingal — Audit blog legacy URL redirects for canonical consolidation
- @mahmoudessam7 — Instrument support ticket tagging for refund-policy mentions (4352-refund-copy-terms-match)
CSS/rendering fix, 75 days. Zero errors, rageclicks dropped from 7/wk to 0-1/wk post-fix. Search page traffic stable at ~33K pvs/wk. No KPI envelope needed for a CSS fix. Recommendation: graduate (confidence 0.88).
93 days, no KPI ever instrumented. City pages (/sublets/) lost 9.2% clicks/day despite +22.6% impressions. Hospital pages (+56.8% clicks) were not the hypothesis target. Recommendation: needs-human-input (confidence 0.35). Question: Were links deployed to city pages, and is the ranking regression acceptable?
81 days. Mobile share did NOT increase (29.4% avg vs 33.9% pre-launch single-week). No target set. CSS polish with no measurable signal at this traffic volume. Recommendation: graduate (confidence 0.70).
79 days. iOS share rose +6.7pp (19.6% to 26.3%). Trend tagged "declining" is inaccurate; recent week was 28 iOS listings (2nd highest post-launch). Recommendation: graduate (confidence 0.82).
58 days. Control deprecated May 8 (46 exposures, frozen). All 3 Pro variants show 24.6-28.1% conversion vs control 17.2%, but control CI is too wide for inference. PostHog returns null for all statistical outputs. Quarterly billing uptake at 4.0% in pro-old-w-quarterly is a real signal. Recommendation: needs-human-input (confidence 0.35).
82 days. Impressions +76% (1,613/day to 2,574/day, accelerating to 3,854/day in last 14d). Clicks +62%. Target of 100 is a smoke placeholder. Legacy WordPress URLs still accumulating clicks (redirect consolidation incomplete). Recommendation: continue (confidence 0.85).
53 days. Refund rate dropped >= 37% (1.52/day to 0.96/day). Conversion rate actually improved 13pp (33% to 46%), contradicting the expected disclosure friction. Secondary KPI (support ticket tagging) never instrumented. Recommendation: continue (confidence 0.70).
94 days. Codebase audit: zero PII references remain. Regression test guards re-introduction. Binary security fix, not a metric. Recommendation: graduate (confidence 0.95).
98 days. Target of 40 indexed pages exceeded (43). Impressions +76%, clicks +62%. Blog-wide CTR 1.1-1.5% (below 3-5% benchmark). 46% of clicks concentrated in 2 Furnished Finder comparison articles. Recommendation: change-target (confidence 0.88).
47 days. All 10 Playwright tests pass. Zero JS exceptions on /need-a-room across 47 days. KPI current_value null due to wrong test path in measurement narrative. Recommendation: graduate (confidence 0.92).
- p1-lister-unread-reminder (5d, @gsingal) — target=1, current=null — too early
- churnkey-subscription-attach (5d) — target=0, current=4 — envelope present, data flowing
- vr-email-null-guard-fixes (5d) — target=0, current=1 — 1 remaining occurrence, monitoring
- pricing-collapse-plans-skeleton (5d) — target=1, current=0 — too early
- last-login-stamp-auth-500-fix (5d) — target=0, current=2 — 2 remaining errors
- lcp-optimization (0d) — target=2500, current=14964 — just launched, LCP far above target
- travel-nurse-landing (0d) — target=20, current=17.09 — just launched, slightly below target
- prebook-duplicate-charges-fix (0d) — target=0, current=7 — just launched, 7 occurrences
- pii-leak-listing-pages-fix (0d) — target=0, current=0 — target met on arrival
- null-location-fix (0d) — target=0, current=3 — 3 remaining occurrences
- admin-user-edit-fix (0d) — target=0, current=0 — target met on arrival
- register-ip-guard (0d) — target=0, current=0 — target met on arrival
- subscription-type-guards (0d) — target=0, current=0 — target met on arrival
- 404-page-fixes (0d) — target=404, current=404 — target met on arrival
- 4259-messaging-logged-out (0d) — target=1, current=0 — too early
- 4338-systemic-checkbox-trait (0d) — target=1, current=0 — too early
- 4316-room-under-review-uncheck (0d) — target=1, current=0 — too early
- 4319-labs-nav-kpi-smoke (0d) — target=1, current=0 — too early
- 4648-keyword-fraud-detector — current=287 vs target=1 (287x) — smoke target, KPI needs recalibration
- 4554-off-platform-contact-detection — current=2038 vs target=5 (408x) — smoke target
- posthog-messaging-instrumentation — current=7234 vs target=1 (7234x) — smoke target
Notable: fraud-lister-only-contact-gate (target=45 direction=lower, current=2026), auth-users-active-on-register-complete (target=0 direction=lower, current=4714), slack-issues-skill-incremental-state (target=0 direction=lower, current=100), ci-test-fixes-round-2 (target=100 direction=higher, current=3).
16 stable launches with continue verdict (scale-collapse threshold active at N>100).
Stable-green IDs
stripe_refund_reconciliation, churnkey-hmac-null-guard, posthog-session-recording-re-enable, geo-radius-listing-price-stats, 4317-breadcrumb-fallback, 3701-room-availability-key-guard, 4617-riskreview-v1-production, listing-active-regression, stripe-link-dual-email-review, mobile-safari-419-handler, messaging-dashboard-timezone, rocket-loader-compat, saveform-ownership-check, edit-listing-500-fix, free-listing-deferred-activation, free-listing-expiration-fix38 launches with broken KPI envelopes (27 with null targets). These need /write-kpis rework or graduation. See B-side audit for the full list and per-launch rework issues.
This week: 5 graduation candidates (map-markers-grey-blocks, email-pii-removal, address-autocomplete-mobile-styling, mobile-address-input-styling, map-z-index-ordering), 1 change-target (blog-content-sitemap), 0 kill candidates, 2 needs-human-input (internal-linking-seo, abc-pro-plan-pricing-v3), 2 continue (blog-migration, 4352-refund-copy-terms-match).