Playbook ingestion — 2026-05-10

Trigger: file watch on toby/strategy/ — toby/strategy/playbook.md created. Source doc: toby/strategy/playbook.md — Toby — Q2 2026 Growth Playbook (last_updated 2026-05-10). Author: presumed toby-strategist (sibling agent that owns the toby/strategy/ sub-folder). Action taken: full ingestion + dashboard rewrite at toby/00-state-of-the-project.md.

What the playbook adds (that didn't exist before in the wiki)#

1. A growth thesis. "The under-pulled lever this quarter is reliability + activation — not acquisition." Backed by the observation that CWS conversion on residual high-intent traffic is ~30% (strong) — so the loss curve is value-realization, not top-of-funnel.
2. An explicit current-loop diagnosis.
   • Primary, declining: CWS organic search (collapsed Oct 8 2025, -83%, never recovered).
   • Secondary, real but ungovernable: word-of-mouth (31% of surveyed users; no in-product mechanic to close the loop).
   • Latent, dormant: the curator content loop (1,848 Free-Tier Archivists; 14,306 active card-share links; 18.6% public-share rate). Only the third is meaningfully under-pulled this quarter.
3. Q2 2026 OKRs — 3 objectives, 9 KRs, all measurable, all with deadlines inside the 7 remaining weeks of Q2.
4. Anti-bets — 6 explicit "we will NOT do this and here's why" calls. The most consequential ones:
   • No pre-announce of AI organize features.
   • No paid acquisition while CWS rank is unrecovered.
   • No Firefox/Safari port this quarter (forces "Phase 2 + reliability OR platform port"; we pick the first).
   • Atlassian/Dia 60-min partnership discovery explicitly deferred past Q2.
5. Red-team pass — three perspectives (sceptical investor / frustrated user / competitor PM) with named residuals that became new open questions for the dashboard.
6. A falsification clause for the thesis itself — if Phase 2 fails AND blank-page hotfix lands AND CWS conversion still falls, the diagnosis is wrong and the bottleneck is structurally upstream.

What the playbook re-confirms (from compass + bets + dashboard)#

• Top-5 priority bets are a strict subset of toby/strategy/bets.md in-flight + proposed.
• The 2026-05-26 Phase 2 decision review (already on the dashboard) is preserved with the original kill criterion.
• The reliability incident is anchor-#1 protection (matches compass).
• AI Q4 deferral preserved.
• Anchor personas (Solo Pro Power Saver, Tenured Free Organizer) preserved as the audience anchors.

What the playbook contradicts (none found, but two near-conflicts to flag)#

1. "Atlassian/Dia is deferred past Q2" (playbook) vs. dashboard's prior framing of it as an immediate-next-step (low-priority but listed). Resolution: dashboard now reflects the playbook position — surface it as "explicitly deferred this quarter; founder-level call after Sept-2026 v3 check-in."
2. Playbook O2 KR1 demands "Phase 2 welcome A/B fully instrumented at canary 5% by 2026-05-19"; bets.md and prior dashboard reflect "decision review 2026-05-26" only. Resolution: not a contradiction — 2026-05-19 is the go-live date that makes the 2026-05-26 decision review possible. Dashboard now surfaces both dates. The 2026-05-19 date raises the silent-slip alarm one week earlier than before.

Changes I made to toby/00-state-of-the-project.md#

• TL;DR: rewrote to reflect that toby/strategy/ now holds the full strategic spine (compass + playbook + bets); playbook is the operational strategy of record for the remaining 7 weeks of Q2.
• Strategic anchor section: restructured into three layers — Identity (compass), Quarter (playbook), Action (bets) — with a "read in order" instruction.
• New section: Q2 2026 OKRs (7 weeks remaining): surfaces all 9 KRs verbatim with deadlines.
• Immediate next steps: re-ordered by playbook ICE + dependency (pricing → reliability → CWS narrative → curator loop → Phase 2). Each item now cites which OKR it serves. Removed prefetchWelcomeExperiment and withExperimentProps slice line items as separate steps — they're now sub-tasks of the Phase 2 canary item, with a hard 2026-05-19 date.
• Atlassian/Dia line: changed from "founder-level decision: take the meeting" to "explicitly deferred this quarter; revisit at founder level after Sept-2026 v3 check-in" — matches playbook anti-bets.
• Roadmap: each timeframe now anchors against named OKR KRs with dates.
• Key decisions: added the 6 Q2 anti-bets as a single new decision; added "top growth lever this quarter is the dormant curator loop, not acquisition channels."
• Open questions: added 4 new entries from playbook red-team and open-questions section (new-tab-override deprecation 2027-2028 risk, team-buyer Slack/LinkedIn canvass, true full-price conversion rate, will reliability hotfix close not_using churn). Updated existing entries to reference the playbook deadlines.
• Doc index: added playbook + bets. Resolves my prior open question ("'bets.md' with unspecified path" — confirmed at toby/strategy/bets.md).
• Recent shipments: backfilled toby/strategy/bets.md publication (I had missed it earlier) and added playbook publication.

What I did NOT do (and why)#

• Did not write into toby/strategy/ — that's the strategist agent's sub-folder per my rules. The "save next to the source" instruction in the orders is overridden by the wiki-scope rule. This ingestion note lives in my own artifacts workspace where it belongs.
• Did not invent any cross-source strategic conclusions. Every line in the rewritten dashboard cites a source path or a commit. The playbook's growth thesis is reproduced as the playbook's growth thesis, not asserted as fact.
• Did not delete prior dashboard content. The "Phase 2 silent slip", "pricing contradiction", and "uncommitted state on main" open questions are still there — the playbook didn't resolve them, it added deadlines to them.

Followup considerations for the operator#

• The playbook implicitly forces the question: who owns each unowned bet? Five of the top-5 bets list "TBD" as owner. Compass anchor #1 (reliability) needs an owner before 2026-05-24 or O1 KR1 misses.
• The 2026-05-19 Phase 2 canary date is now the earliest visible alarm bell for "is Jad shipping?" — earlier than the 2026-05-26 decision review. The dashboard's Open Questions surface this.
• The team-buyer LinkedIn canvass (open question from playbook) is operator-led and zero-engineering. Worth surfacing in a future check-in.