Skip to content
Back to skills

Gtm Alignment

ASecurity

Align phases with GTM strategy, persona tiers, and KPI targets

  • 342 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added September 1, 2026
toolsgoapidatabase

Works with

  • api
  • mcp

Security analysis

A100/100

Scanned September 1, 2026

npx -y skills add WellApp-ai/Well --skill gtm-alignment --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Gtm Alignment?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Gtm Alignment
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/wellapp-ai-gtm-alignment/badge)](https://www.skillsdirectory.com/skills/wellapp-ai-gtm-alignment)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: gtm-alignment
description: Align phases with GTM strategy, persona tiers, and KPI targets
---

# GTM Alignment Skill

Ensure implementation phasing aligns with go-to-market strategy by validating against persona tiers and estimating business value from KPI database.

**CRITICAL:** This skill MUST fetch real data from Notion. Do NOT make up personas or GTM tiers.

## When to Use

- During Ask mode Phase 2 (CONVERGE), after dependency-mapping
- When prioritizing between equally-risky slices
- Before finalizing phase order

## Notion Database IDs

| Database | ID | Purpose |
|----------|-----|---------|
| Persona DB | `2b1c4d5e-7bea-80ee-85c2-fb96dfdcf98b` | Persona tiers, RICE scores, characteristics |
| GTM Strategy | `2b2c4d5e-7bea-809d-8c49-cc6b443738df` | Positioning, messaging, segments |

**Note:** If `API-query-data-source` fails, use `API-post-search` with query "Persona" as fallback.

## Instructions

### Phase 1: Fetch Persona Tiers (REQUIRED)

**Primary method - Query database:**
```
CallMcpTool:
  server: user-notion
  toolName: API-query-data-source
  arguments: {"data_source_id": "2b1c4d5e-7bea-80ee-85c2-fb96dfdcf98b"}
```

**Fallback method - Search (if query fails):**
```
CallMcpTool:
  server: user-notion
  toolName: API-post-search
  arguments: {"query": "Persona"}
```

**STOP if BOTH methods fail.** Report error and do not proceed with made-up data.

**Extract from Persona DB:**
- Persona name (title)
- RICE score (for prioritization)
- Goals, Pain points, Background
- Market segments

**Build Tier Priority Table (sorted by RICE score):**

| Tier | Persona | RICE Score | Key Traits |
|------|---------|------------|------------|
| T1 | [Highest RICE from Notion] | [Score] | [Goals/Pain points] |
| T2 | [Second RICE from Notion] | [Score] | [Goals/Pain points] |
| T3 | [Third RICE from Notion] | [Score] | [Goals/Pain points] |

**Evidence required:** Show persona names and RICE scores from Notion

### Phase 2: Fetch GTM Strategy (REQUIRED)

**Primary method - Query database:**
```
CallMcpTool:
  server: user-notion
  toolName: API-query-data-source
  arguments: {"data_source_id": "2b2c4d5e-7bea-809d-8c49-cc6b443738df"}
```

**Fallback method - Search (if query fails):**
```
CallMcpTool:
  server: user-notion
  toolName: API-post-search
  arguments: {"query": "GTM Strategy"}
```

**STOP if BOTH methods fail.** Report error and do not proceed with made-up data.

**Extract from GTM Strategy:**
- Current positioning statement
- Key differentiators vs. competitors
- Priority segments/verticals
- Messaging themes

**Evidence required:** Show positioning statement excerpt from Notion

### Phase 3: Map Slices to Personas

Using the ACTUAL persona data from Phase 1, map each slice to the persona it serves:

| Slice | Serves Persona | Tier | Why |
|-------|----------------|------|-----|
| [Slice name] | [Persona from Notion] | T1/T2/T3 | [Rationale] |

### Phase 4: Calculate GTM Score

```
GTM Score = (TierPriority x 3) + AnnouncementPotential

Where:
- TierPriority: T1=3, T2=2, T3=1, None=0
- AnnouncementPotential: Yes=2, Partial=1, No=0
```

### Phase 4: Calculate GTM Score

```
GTM Score = (TierPriority x 3) + ValueScore + AnnouncementPotential

Where:
- TierPriority: T1=3, T2=2, T3=1, None=0
- ValueScore: Critical=3, High=2, Medium=1, Low=0
- AnnouncementPotential: Yes=2, Partial=1, No=0
```

### Phase 5: Map Phases to Personas

| Phase | Serves Tier | Persona | Features Included | Can Announce? |
|-------|-------------|---------|-------------------|---------------|
| 1 | T1 | [Primary] | [Slice list] | Yes/No |
| 2 | T1, T2 | [Primary, Secondary] | [Slice list] | Yes/No |
| 3 | All | All personas | [Slice list] | Yes (Full Launch) |

**Rule:** Each phase should fully serve at least one persona tier.

### Phase 6: Reconcile with Risk Order

If GTM Score conflicts with Risk Score (from dependency-mapping):

| Scenario | Resolution |
|----------|------------|
| High GTM + High Risk | Split slice into MVP + Full (reduce risk, keep GTM priority) |
| High GTM + Low Risk | Prioritize (move earlier in sequence) |
| Low GTM + Low Risk | Keep in risk-based position |
| Low GTM + High Risk | Defer (move later in sequence) |

**Document conflicts for checkpoint:**
```
CONFLICT DETECTED:
- Risk order: #1.1 -> #1.2 -> #2.1 -> #2.2
- GTM order: #2.2 -> #1.1 -> #1.2 -> #2.1 (Invite flow is T1 priority)

RESOLUTION OPTIONS:
- Option A: Accept risk, prioritize GTM (ship #2.2 first)
- Option B: Split #2.2 into MVP + Full (ship #2.2-MVP in Phase 1)
- Option C: Ship #1.1 first, fast-follow with #2.2 (compromise)
```

## Output Format

Include **evidence of Notion fetch** in output:

```markdown
## GTM Alignment

### Data Sources (REQUIRED)

| Source | Status | Evidence |
|--------|--------|----------|
| Personas DB | ✓ Fetched | [N] records from `2b2c4d5e...` |
| GTM Strategy DB | ✓ Fetched | Positioning: "[first 50 chars...]" |

### Persona Tiers (from Notion)

| Tier | Persona | Characteristics |
|------|---------|-----------------|
| T1 | [From Notion] | [From Notion] |
| T2 | [From Notion] | [From Notion] |

### Strategy Fit
- Positioning: [From Notion GTM Strategy]
- Messaging: [Key message this enables]
- Differentiator: [Competitive advantage]

### Persona Coverage

| Phase | Serves Tier | Persona | Can Announce? |
|-------|-------------|---------|---------------|
| 1 | T1 | [Persona from Notion] | Yes/No |
| 2 | T1, T2 | [Personas from Notion] | Yes/No |

### Conflicts & Resolutions

| Conflict | Risk Order | GTM Order | Resolution |
|----------|------------|-----------|------------|
| [Description] | [Sequence] | [Sequence] | [Chosen option] |
```

## Validation Checkpoint

Before proceeding, verify:

- [ ] Personas DB queried successfully (not made up)
- [ ] GTM Strategy DB queried successfully (not made up)
- [ ] Each slice mapped to a real persona from Notion
- [ ] Evidence shown in output (record counts, positioning excerpt)

**If any query fails:** STOP and report the error. Do NOT proceed with invented data.

## Invocation

Invoke manually with "use gtm-alignment skill" or follow Ask mode Phase 2 (CONVERGE) which references this skill.

## Related Skills

- `dependency-mapping` - Run before GTM alignment
- `problem-framing` - Provides persona context
- `phasing` - Uses GTM scores for final ordering

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.