Unclassified

45 — Onboarding Playbook

Status: CANONICAL · Commercial OS
Owner: Founder (CS)
Last verified: 2026-07-21

Purpose

Days 0–30, minute-by-minute where it matters. Onboarding has one job: make the product's promise physically real before the optimism window closes. The promise: first value same-day; the ritual replaced in week one; activation by day 30.

Target audience: CS operator; sales (to promise exactly this and no more).

1. Design Principles

Hydration before ceremony. Tools connect in the kickoff meeting itself; data flows while we talk. No "phase 1: alignment workshops."
The first morning is the product. Every seat's day-1 experience is the workbench opening on their accounts with the three beats. We engineer that moment; nothing else matters until it lands.
Time-to-first-value is designed in, not hoped for (activation doctrine): the workbench opens after minutes-to-hours of initial hydration; deep history back-fills in the background.
Baselines or it didn't happen. The metrics that win the renewal are captured in week one (28 §4).

2. The 30-Day Arc

Day 0 — Kickoff (60 min, all POV/paid seats + P2)

Min	Beat
0–10	Frame: "By the end of this call your tools are connected and hydrating. By tomorrow morning, your workbench knows your accounts." Success plan on screen — the 3 committed outcomes with numbers (28), inherited from the sales handoff (37 Play 6)
10–35	Live connection ceremony: CRM first (HubSpot/Salesforce OAuth — minutes), then support/billing/comms. Each connector: "connected → hydrating" visible on screen. This is the demo of "no IT project"
35–50	Baseline capture (while hydrating): each seat's H (self-reported morning minutes), QBR prep hours, last-surprise story. Recorded in the success plan
50–60	The contract: "Tomorrow 9am, open your workbench before anything else. Day 2 I'll call to hear what you saw. Week 1 we do first-morning reviews. Day 30 we measure against these baselines."

Day 1 — The First Morning (engineered, not left to chance)

Pre-flight (us, before their 9am): hydration status per connector · entity resolution sanity (their top-10 accounts assembled correctly) · at least one legitimate what changed present · signals thresholds sane for their data shape. If the workbench isn't ready, we call before they open it — a broken first morning costs the whole optimism window.

Day 2 — First-Morning Debrief (15 min/team)

"Open your workbench. Narrate what you see." Listen for: recognition ("that's my account") · surprise ("it already knows X") · gaps (missing tool, wrong mapping — fix same-day). First Ask-Your-Twin together; first Store-in-Spine together. The four controls taught in their data, never in a slide.

Week 1 — The Ritual Replacement

Each seat: morning-workbench habit (the email/Slack nudge: "What changed for your accounts today?")
First Assign: every seat delegates one real prep (meeting brief, account summary) and reviews the evidence
First Approve: at least one governed action per seat — the trust threshold crossed personally
Us: daily 15-min office hours; connector/mapping fixes < 24h; week-1 time sample (morning minutes, measured again)

Weeks 2–3 — Depth and the First Catch

Signal review: are surfaced signals real to them? Tune thresholds with their feedback (the noisy-v1 lesson from the build plan: iterate openly)
TruthLayer habit: approve/reject memory proposals in the weekly rhythm; the org's verified memory starts compounding
Target: first documented catch by day 21 (44 play if quiet)
Manager view for P2: team adoption + early time-deltas

Day 30 — Activation Review (P2 + team lead, 30 min)

The scorecard, honestly: activation % vs. 70% target · morning minutes: baseline → now · approved actions/seat/week · catches with evidence · blockers with owners. Then: success plan confirmed/adjusted; cadence set (44 §3); expansion seeds noted silently (46).

3. Activation Definition (canonical — 75)

A seat is activated when, in a single week, it has ≥ 5 approved Twin actions and ≥ 4 workbench-first mornings. An account is activated at ≥ 70% of seats activated by day 30.

4. Failure Modes & Countermeasures

Failure	Countermeasure
Connector stall (missing admin/OAuth rights)	Pre-kickoff checklist to champion: admin availability confirmed for the call
"Another tab" syndrome (workbench beside the old ritual)	Manager models it: team standup runs from the workbench, week 1
Data-quality shock ("that mapping is wrong")	Frame at kickoff: "You'll find mapping gaps week one — that's normal; we fix same-day. The provenance panel shows you exactly what came from where"
Twin over/under-surfacing (noise or silence)	Threshold tuning session week 2; honesty about iteration (v1 noisy → v4 good)
Baseline skipped in kickoff excitement	Success plan template has required fields; day-30 review literally cannot run without them — structural forcing function

KPIs: same-day hydration rate 100% · day-1 workbench readiness 100% (pre-flight compliance) · week-1 ritual replacement (≥ 4 workbench-first mornings/seat) ≥ 60% of seats · day-30 activation ≥ 70% · first catch ≤ 21 days ≥ 70% of accounts · time-to-first-value (kickoff → first approved action) ≤ 48h median.