L’interface est localisée ; les présentations et documents des Skills sont traduits automatiquement par IA dans votre langue. Le code, les commandes, chemins et termes techniques restent en version originale, et le document de l’auteur fait foi.
Explorer les Skills
Productivité · SKILL.md

workflow-gtm

workflow-gtm

Take a shipped repo to market: plan-gtm audit and interview, approval, then measure, message, activate, get found, launch, and a weekly loop. Use when "go to market", "grow users", or "increase traffic and visibility". Strategy only → plan-gtm.

Independent GitHub snapshot · Automated classification · Original documentation governs

Documentation de la Skill

GitHub

name: workflow-gtm description: > Take a shipped repo to market: plan-gtm audit and interview, approval, then measure, message, activate, get found, launch, and a weekly loop. Use when "go to market", "grow users", or "increase traffic and visibility". Strategy only → plan-gtm. license: MIT metadata: chain: "plan-gtm audit-analytics enhance-web-conversion audit-registry-listing enhance-readme enhance-onboarding enhance-lifecycle-email enhance-web-seo docs-comparison-pages plan-aeo-readiness docs-launch-kit enhance-growth-loops iterate-gtm-weekly"

workflow-gtm — From shipped to found, used, and paid

Degree of freedom: MIXED. Skip/condition judgment per step [HIGH freedom]; step order, the approval gate after Step 1, and the scorecard [LOW freedom — run exactly].

The go-to-market sequence for a product that already works. Plan first, get the founder's approval, then run the apply skills in the order that compounds: measure → message → activate → be found → launch → loop.

Technical launch readiness (SEO meta, PWA, bundle, quality gate) is workflow-launch-ready, not this. Run it before Step 5 when the app has never shipped.

How to reason

  1. Scope — shipped product, real users possible today; not a PRD
  2. Gate — plan-gtm.md approved phase by phase; nothing applied before
  3. Sweep — which steps apply this quarter; honest skips named
  4. Scorecard — funnel table before and after; weekly loop scheduled

Worked example

Scope: Next.js + Supabase kakeibo app, live, ~200 signups/month, no paid tier yet. Gate: plan-gtm.md approved Phases 1–3; Phase 5 (monetize) parked until activation ≥ 30%. Sweep: analytics → hero → onboarding → SEO/AEO → launch kit for v2.1; skip pricing. Scorecard: activation unmeasured → 22% → 34% after onboarding; weekly loop on Mondays via iterate-gtm-weekly.

Self-critique before reporting

  • Plan approved — no Step 2+ skill ran before the founder approved its phase
  • Measured first — the funnel table had real cells (or audit-analytics added events) before copy or UX changed
  • Skips named — every skipped step has a one-line reason in the scorecard
  • No promised outcomes — the scorecard reports numbers, not "should grow"
  • Right owner — PRD → design-prd; technical launch sweep → workflow-launch-ready

Sequence [LOW freedom — run exactly]

1. PLAN      → plan-gtm                 (inventory, interview, plan-gtm.md)  ── STOP for approval
2. MEASURE   → audit-analytics          (Visit→Signup→Activated→Paid events, consent-gated)
3. MESSAGE   → enhance-web-conversion   (hero, CTA, proof; pricing only if Phase 5 approved)
             → enhance-readme           (README as landing page, when the repo is the product;
                                         audit-registry-listing first for the findings)
4. ACTIVATE  → enhance-onboarding       (activation event, cut steps, templates, checklist)
             → enhance-lifecycle-email  (activation nudges keyed to the same events)
5. BE FOUND  → enhance-web-seo          (meta, sitemap, schema)
             → docs-comparison-pages    ("X vs Y" / alternatives pages, ≤5, dated facts)
             → plan-aeo-readiness       (AI-engine citation plan; approve, then enhance-web-seo)
6. LAUNCH    → docs-launch-kit          (per-channel kit for this release)
             → enhance-growth-loops     (badge / share / referral, one K per loop)
7. LOOP      → iterate-gtm-weekly       (weekly: funnel by source, one experiment, one post)

Steps 3–6 run only for phases the founder approved. Mobile store products add plan-aso beside Step 5. Phase 5 (monetize) runs plan-pricing → approval → enhance-web-conversion (pricing) + enhance-lifecycle-email (trial expiry).


Step 1: Plan (read plan-gtm) [LOW freedom — hand off, then stop]

Read the plan-gtm skill and follow it.

Ends the turn with plan-gtm.md. Wait for phase approval.

Step 2: Measure (read audit-analytics) [HIGH freedom]

Read the audit-analytics skill and follow it.

Goal: every funnel cell measurable, consent-gated, one taxonomy. Include the activated event plan-gtm defined.

Step 3: Message (read enhance-web-conversion, enhance-readme) [HIGH freedom]

Read the enhance-web-conversion skill and follow it. For repo-as-product, run audit-registry-listing for findings, then enhance-readme.

Hero carries the positioning statement; one CTA; real proof.

Step 4: Activate (read enhance-onboarding, enhance-lifecycle-email) [HIGH freedom]

Read the enhance-onboarding skill and follow it. Then read enhance-lifecycle-email for the activation nudges keyed to the same events.

Report steps before/after and the activated event firing.

Step 5: Be found (read enhance-web-seo, docs-comparison-pages, plan-aeo-readiness) [HIGH freedom]

Read enhance-web-seo and follow it. Read docs-comparison-pages for the ≤5 bottom-funnel pages. Then read plan-aeo-readiness; its plan is approved before its fixes run.

Step 6: Launch (read docs-launch-kit, enhance-growth-loops) [LOW freedom — hand off]

Read the docs-launch-kit skill and follow it. If the plan named a loop, read enhance-growth-loops.

One kit per release; UTMs on every link; one K per loop.

Step 7: Loop (read iterate-gtm-weekly) [LOW freedom — hand off]

Read the iterate-gtm-weekly skill and follow it.

Weekly review, five lines: funnel by source · activation cohort · top drop-off step · one experiment shipped · one launch or post. Production defects stay on iterate-post-launch.


Scorecard output [LOW freedom — run exactly]

## GTM Scorecard — <product> — <date>

### Decision log (from plan-gtm.md)
Goal metric · ICP · Model · Budget

### Funnel
| Stage | Before | After | Source |
|-------|--------|-------|--------|
| Visit | | | |
| Signup | | | |
| Activated | | | |
| Habit | | | |
| Paid | | | |
| Refer | | | |

### Steps
| Step | Status | Skill | Notes / skip reason |
|------|--------|-------|---------------------|
| 1 Plan | approved phases: … | `plan-gtm` | |
| 2 Measure | ✅ / ⚠️ / skipped | `audit-analytics` | |
| 3 Message | | `enhance-web-conversion` | |
| 4 Activate | | `enhance-onboarding` | |
| 5 Be found | | `enhance-web-seo`, `plan-aeo-readiness` | |
| 6 Launch | | `docs-launch-kit` | |
| 7 Loop | scheduled <day> | `iterate-gtm-weekly` | |

### Next week's one fix
Skills enregistrées