Acquisition.md
← Playbooks
Email · 28 min

Lifecycle Blueprint

Onboarding, activation, retention and win-back flows.

Lifecycle isn't a series of emails. It's a series of moments. Design the moments; the emails are just delivery.

tl;dr
  • Lifecycle wins in 2026 come from Attio + Clay + 11x, not 14-person lifecycle teams.
  • Onboarding should trigger in <5 minutes, or PostHog and Common Room have already failed you.
  • Activation is a 3-event rule: signup, value event, and team invite within 7 days.
  • Retention is 1:1 sequencing with Clay scores, not one HubSpot blast to 200k contacts.
  • Win-back is dead mail-merge unless Relevance AI, Lindy, and Claude 4.5 Sonnet personalise it.
Action plan
Action plan0/3 · 0%
Progress saved locally. Sign in to sync
Week one
  1. 01Day 1: Map 1 lifecycle schema in the CRM with 4 stages: new, activated, retained, win-back.
  2. 02Day 2: Build 1 enrichment workflow and verify 90%+ coverage on a 100-record sample.
  3. 03Day 3: Ship 8 product events and confirm they write back to the CRM.
  4. 04Day 4: Draft 3 onboarding paths and test them on a preview page.
  5. 05Day 5: Write 2 warehouse SQL queries that expose activation gaps by cohort.
  6. 06Day 6: Launch 1 outbound sequence for lagging activations and measure reply rate.
  7. 07Day 7: Review 1 cohort and cut every lifecycle step with <5% lift.

Onboarding: move from signup to first value in 1 session

Onboarding in 2026 is a 4-step system: detect intent, enrich instantly, route by use case, and trigger a human fallback only when the agent stalls. Clay should score the lead in under 15 minutes; PostHog should confirm the first 2 events; Attio should advance lifecycle stage automatically. 11x or Lindy handles the first email, while Claude 4.5 Sonnet drafts the variant if the persona is ambiguous. Framer + Relume pages should mirror the lifecycle promise. If your onboarding is still a single HubSpot drip, you are paying for noise, not conversion.

  1. 01

    Detect and enrich in 15 minutes

    Use Clay to enrich company size, role, and intent, then push the score into Attio. Trigger.dev runs the flow when signup lands, and PostHog stores the source event.

    If {{signup_source}} = 'demo' and Clay score > 70, then send 11x email #1 and set Attio stage to 'Activated-pending'.
  2. 02

    Send 1 persona-specific path

    Use Relevance AI or Lindy to choose 1 of 3 onboarding paths: founder, operator, or team lead. Claude 4.5 Sonnet writes the subject line and CTA from the same prompt.

    Prompt: 'Write a 48-char subject for {{persona}} using {{job_to_be_done}} and 1 proof point.'
  3. 03

    Close the loop with product events

    PostHog watches for first key action, Common Room flags team adoption, and n8n updates Attio when the user hits activation. No manual CSVs.

    If event='shared_workspace' within 7 days, then mark lifecycle='activated' and stop onboarding sequence.
IfThen
Clay enrichment confidence is <0.75Use Apollo AI or Common Room as fallback and keep the path generic for 1 send.
PostHog shows 2 sessions but 0 key events after 24 hoursFire a 11x human-style check-in and route to a 1-click setup guide in Framer.

Activation: define value with 3 events, not vibes

Activation needs a hard threshold in Attio and PostHog, or every dashboard becomes fiction. The best 2026 rule is 3 events in 7 days: one setup event, one collaboration event, and one value event. Common Room spots product-led signals at account level, while Clay enriches likely champions for the next nudge. If you need SQL, use Cursor or Claude Code against PostHog and Attio schemas, then write the cohort back via n8n. The winning tactic is a rule engine, not a newsletter. HubSpot AI can host the ops, but it should never own the decision.

  1. 01

    Choose 3 activation events

    Pick 1 setup event, 1 team event, and 1 value event in PostHog. Keep the rule visible in Attio and Common Room so every operator can audit it.

    activation = signup + invite_sent + dashboard_shared within 7 days.
  2. 02

    Query the laggards

    Use Cursor or Claude Code to run RevOps SQL on event gaps, then segment by persona, source, and company size. Feed the segment to Clay for enrichment.

    Prompt: 'Find accounts with signup but no invite_sent after 72 hours; return 3 reasons and 1 next action.'
  3. 03

    Trigger the right nudge

    11x or Lindy sends the nudge, and n8n updates Attio stage plus a PostHog property. Keep the CTA to 1 action only.

    If no value event by day 5, then send 'Need 2 minutes to set this up?' and link the exact screen.
IfThen
Activation rate is below 35% for 14 daysDelete 1 onboarding step before adding another email; friction beats volume every time.
A cohort activates but does not invite teammatesUse Common Room to identify the champion and send a team-expansion prompt within 24 hours.

Retention: keep the product sticky with behavioural lifecycle loops

Retention in 2026 is a 2-layer system: behavioural triggers from PostHog and account context from Clay plus Common Room. The old play is a monthly newsletter; the winner is a usage-based loop that reacts to drops in frequency, feature depth, or team breadth. Attio should hold the lifecycle stage, while n8n routes the right sequence based on account health. If your emails do not reference a live event, they are useless. Use Claude 4.5 Sonnet to rewrite the message from the user’s last action, then let 11x handle the human-like follow-up when the score crosses a threshold.

  1. 01

    Build 1 health score

    Combine frequency, breadth, and recency from PostHog into 1 account score. Store it in Attio and expose it in Common Room for CS and RevOps.

    health_score = 0.4*frequency + 0.3*breadth + 0.3*recency.
  2. 02

    Send event-led retention emails

    Trigger emails only when a user misses a habit, hits a ceiling, or discovers a new feature. Clay enriches the account and Claude 4.5 Sonnet writes the one-line reason.

    Prompt: 'Write a retention email for {{account}} using the last 2 events and 1 upgrade angle.'
  3. 03

    Escalate by account risk

    If the account is high-value and usage drops 30%+, fire an 11x agent plus an Attio task for a human owner. No manual triage queue.

    If MRR > 500 and health_score < 0.4, then create owner task and send 1 re-engagement email.
IfThen
A user opens 3 emails but stops using the productStop emailing features and send a single usage-reset CTA tied to the missing event.
Common Room detects team adoption on 1 seat but not 3Switch from product education to collaboration proof and expansion framing.

Win-back: reactivation is an agentic workflow, not a blast

Win-back should start in Attio the moment churn is marked, then branch by reason, value, and likelihood to return. Clay enriches the dormant account again, Common Room checks for fresh intent, and Relevance AI or Lindy chooses the message style. Claude 4.5 Sonnet should draft 3 variants from 1 prompt, and 11x can send the most human version after a failed trial or a contract lapse. If the churn reason is price, send product proof; if it is adoption, send a 2-step restart. Generic 'we miss you' mail is dead in 2026.

  1. 01

    Classify churn reason fast

    Use Clay, Attio notes, and PostHog history to label churn as price, adoption, timing, or missing feature. Push that label into n8n and the email tool.

    churn_reason = 'adoption' when usage fell 40% before cancel.
  2. 02

    Generate 3 reactivation variants

    Use Claude 4.5 Sonnet to draft 3 short emails, and let Relevance AI choose the best angle by persona. Keep each email tied to 1 win-back goal.

    Prompt: 'Write 3 win-back emails for {{reason}} with 1 CTA and 1 proof point each.'
  3. 03

    Escalate only high-probability accounts

    Use Common Room and Clay to find accounts with fresh intent or new hiring signals, then let 11x handle the personal follow-up.

    If new job_posting_count > 3 and churn_reason='timing', then send 'ready to restart?' from 11x.
IfThen
No reply after 2 win-back touchesStop the sequence for 90 days and move the account into an intent-only watchlist.
Clay finds a new champion at the same churned accountRestart with a fresh use-case email, not the old cancellation narrative.
Pitfalls

What goes wrong

Mistake

Using one generic lifecycle journey for 5 personas

Fix

Split by role in Clay and route 3 paths in Lindy or Relevance AI.

Mistake

Letting HubSpot AI own logic and Attio own nothing

Fix

Keep Attio as the source of truth and write back every stage change in n8n.

Mistake

Measuring opens instead of product events

Fix

Use PostHog and Common Room as the trigger layer; opens are only a secondary signal.

Mistake

Sending retention mail without a live usage reason

Fix

Require 1 event, 1 missed habit, or 1 expansion signal before every send.

Mistake

Building fragile automation in Zapier for AI-heavy workflows

Fix

Use n8n, Trigger.dev, or Inngest so agent retries and branching do not break on 1 webhook.

FAQ
Do I need 1 lifecycle tool or 5?▾

Use 5 primitives, not 5 platforms: Attio for truth, Clay for enrichment, PostHog for events, n8n for orchestration, and 11x or Lindy for sends. HubSpot AI is fine for marketing-heavy teams, but it should not hold the decision layer. If one tool owns logic, you will debug it for 6 months.

What is the fastest win in week 1?▾

Cut onboarding latency below 5 minutes. Build 1 Trigger.dev flow, 1 Clay enrichment pass, and 1 Attio stage update. Then send only 1 persona-specific email from 11x. That single change usually beats 4 extra nurture emails because speed is the cheapest conversion lever in 2026.

How do I define activation without arguing for 3 weeks?▾

Pick 3 events in PostHog, then make 1 of them collaborative. The clean rule is signup, value action, and teammate invite within 7 days. If your product needs more than 3 events, you are probably measuring exploration, not activation. Use Cursor or Claude Code to validate the cohort before freezing the rule.

When should I use AI agents instead of a normal workflow?▾

Use AI agents when the next step needs judgement: persona selection, churn reason classification, or message rewriting. Use n8n for deterministic routing and 11x, Lindy, or Relevance AI for the variable branch. If the branch count is under 2 and the copy is fixed, an agent is waste.

What is the right win-back sequence length?▾

Use 3 touches over 10 days, then stop for 90 days. The first touch should reference the churn reason, the second should add a proof point, and the third should offer 1 restart path. Anything longer becomes spam, and anything shorter misses the reactivation window visible in Common Room and Clay.

Get one system like this in your inbox each week

No fluff, no upsell. Unsubscribe anytime.