A tiered GTM launch process. Synthesizes Yi Lin Pei's 4-step launch flow (Strategic Readiness → Positioning/Messaging → External Promotion → Internal Enablement) with a deliverables matrix sized by tier.
When to use
Any product, feature, integration, or repositioning launch. Tier the launch first; resource allocation flows from there.
Tier classification
- T0, Company-defining launch. New category, repositioning, flagship product. Full multi-channel, full enablement, executive comms.
- T1, Major feature. Full asset suite, sales enablement, multi-channel external promo.
- T2, Notable feature. Lighter assets, in-app + email + release notes.
- T3, Maintenance / minor improvement. Release notes only.
The most common failure mode is treating every launch as T0.
Five phases
- Planning. Tier classification, stakeholder map (RACI), readiness checklist (positioning lock, feature-complete, baseline capture, attribution setup, in-app activation verification, enablement gate), evidence freshness check (run signals before locking messaging).
- Enablement. Sales/CS training BEFORE external launch. Persona-specific demo paths (3–5 for T1+ with 3+ ICP segments). UX-change FAQ. Rollback notification standby. Mandatory 48h-prior recorded webinar dry-run.
- Launch. Cross-functional sync. Asset production. Coordinated go-live across channels.
- Post-launch. Soft-launch playbook with phase gates. Customer interviews. Support-load monitoring. Asset refresh. Substrate update list named at kick-off.
- Measurement and retro. Three-tier outcome framework: conversion, adoption, revenue. Phased gate progression. Documented learnings.
Deliverables matrix (size by tier)
External: blog post, landing page, customer email, in-app banner/announcement, sales note, comparison page, demo video, social posts, paid amplification, partner co-marketing.
Internal: sales deck, battlecard, FAQ, CS playbook, training session, dry-run, exec briefing, comms cadence (monthly sync → kick-off → working group → roadshow → retro → monthly recaps).
Customer comms mechanics
- Sender ≠ reply-to. Two separate decisions.
- Reply-to is bucket-keyed: Strategic CSM, Scaled CSM, or scaled alias. Never CEO@.
- When the audience splits on state (registered vs not-registered, in-progress vs rejected), send distinct emails to distinct lists. Do not A/B test across state.
- One banner per state. Registration, in-progress, rejected, pre-flight all need distinct copy + CTA.
Quality gates
- Tier classification justified against scope and customer impact.
- Every deliverable has owner, due date, collaborators. No orphans.
- Messaging derives from positioning doc, not invented for the launch.
- Internal enablement complete before external go-live.
- KPIs measurable and tier-appropriate. No vanity metrics.
- Pre-launch baseline captured 48h prior; attribution live 24h prior; in-app flows manually verified within 2h of phase launch.
- Customer-grade output gate passes: no internal-process language; reads as if the customer wrote it to themselves.
- Substrate update list named: what does this launch promote back into kill-list, voice rules, battlecards, templates?
- Cross-functional buy-in documented.
- Historical outcomes from similar launches referenced to set realistic targets.
Common failure modes
- Treating every launch as T0; spreading resources thin.
- Skipping the positioning step; messaging invented ad hoc.
- No post-launch plan. Go-live treated as finish line.
- Assets shipped but never activated (in-app tours toggled off).
- Unattributable outcomes; no tracking plan.
- One person blocking video/design escalates the whole timeline.
- Launch cancelled after content created, insufficient pre-launch scope alignment.
- LP with views and zero conversions (missing CTA / form).
- PMM brought in too late, PM excluded from GTM decisions.
- Overloading launches with channels. Focus on 1–3 effective channels per tier.
- Stale-positioning launch shipping messaging from a doc written before recent customer evidence.
- Substrate not updated post-launch, same mistakes recur.
- Voice violations shipped (em-dashes, "agents" vs "reps", forced line breaks, mixed-customer pullquote+stats).
- Marketing duplicating instrumentation product already runs. Read existing dashboards first; 1-line ask if a metric is missing.
- Help-article precedence ignored when two articles cover the same topic. Lead with the recently-updated one; verify via Wayback Machine.
- Asset scope creep mid-launch from feature scope change. Require written feature-lock before asset production.
- Enablement shipped AFTER external launch. Hard gate.
- No pre-launch baseline. Post-launch metrics uninterpretable.
- UX change without "where did X go?" FAQ.
Outputs
- Tier classification with rationale.
- Deliverables checklist with owners + dates.
- External channel strategy.
- Internal enablement plan.
- Internal comms cadence.
- KPI + reporting schedule.
- Post-launch plan + substrate update list.