Beachhead Segment

Beachhead-Segment — Skill

How This Works

Identifies the highest-probability segment to dominate first — before expanding. This is the wedge. Everything else (positioning, GTM strategy, proof points) follows from getting this right.

The skill runs in 7 steps:

Step 0 — Load brain context (ICP, positioning, competitive landscape, proof points) and, if the user maintains /context/meta-patterns.md, guardrails logged there.

Step 1 — Identify candidates: Name 2–5 segments or ask user to decompose current ICP.

Step 2 — Score each segment on four dimensions: Burning Pain, Willingness to Pay, Winnability, Referral Potential.

Step 3 — Apply blocking gates: Pain floor (≥3), Winnability floor (≥3), assumption density check.

Step 4 — Recommend beachhead with expansion pathway, 90-day activation plan, and specific rejection reasons for every eliminated segment.

Step 5 — Update brain Section 2 with confirmed beachhead (on user confirmation).

Step 6 — Learning Close: log the session to /context/skill-sessions.md.

Correction (2026-08-24): this summary previously listed 8 entries (Step 0 through Step 7) under "runs in 7 steps," and named a standalone "Step 5 — Pressure-test eliminated segments" that never existed as its own section in the body — the body always went straight from Step 4 (Recommend) to what it labeled Step 5 (Update Brain). Rejection reasons for eliminated segments were already produced inside Step 4's own output template ("Why not Segment B/C" lines and the Eliminated Segments table) — that was real, just mislabeled as a separate step. The step count and numbering above now match the body exactly: 7 steps, numbered 0–6.

Trigger

• When: Choosing which customer segment to focus on first, before scaling GTM investment across multiple segments at once — narrowing a broad ICP down to the first wedge.
• Not for: Full ICP definition from scratch → use product-marketing-context. Launch tier assignment once the beachhead is already confirmed → use go-to-market-strategy. Mapping the buying committee inside a confirmed beachhead → use buyer-personas. Messaging for a confirmed beachhead → use positioning-messaging.
• Example prompts:
  • "Which segment should we focus on first?"
  • "Our ICP is too broad — help me narrow it"
  • "Pick a beachhead for us"
  • "Where do we win first?"
  • "What's our wedge?"

Inputs

• Args: 2–5 candidate segments, or a description of the current customer base to decompose. Free format.
• Defaults: If only one segment is named, this skill challenges it before scoring — a single segment isn't a choice, it's an assumption (Step 1). Never scores fewer than 2 candidates.
• Context keys:
  • /foundation/brain.md — recommended. Sections 2 (ICP), 3 (Positioning), 4 (Competitive), 5 (Proof Points) loaded silently at Step 0.
  • /context/meta-patterns.md — optional; recurring patterns the user has logged from prior beachhead decisions.
  • Brain contract: Reads Sections 2, 3, 4, 5. Writes Section 2 — the confirmed beachhead, expansion pathway, and scores (Step 5), only after explicit user confirmation. Never writes to any other section.

Pre-flight

• Load /foundation/brain.md Sections 2, 3, 4, 5 if it exists — see Step 0 for the full sequence.
• Load /context/meta-patterns.md if it exists, and surface any guardrail that has fired 2+ times in prior beachhead decisions — see Step 0.
• Hard block: if /foundation/brain.md is absent or Section 2 (ICP) is empty, stop and surface the message in Step 0's Gate check before proceeding — beachhead scoring without ICP, positioning, and competitive context produces guesswork, not a defensible recommendation.

Steps

Step 0 — Pre-Flight: Load Context & Surface Guardrails

Before intake, load:

• Brain context (Sections 2, 3, 4, 5): ICP, positioning, competitive landscape, proof points — these anchor all four-dimension scoring
• Guardrails from /context/meta-patterns.md, if that file exists in the user's workspace: if a pattern has actually fired 2+ times in prior beachhead decisions logged there, surface it now

Surface guardrails like this:

🔁 PATTERN FROM PRIOR BEACHHEAD DECISIONS

I've seen [pattern description] in 2 prior sessions.
Examples: [specific segments or outcomes]

Quick check: Does this apply to your candidates?
- If YES → We'll watch for this during scoring
- If NO → Let's flag it if it emerges

You can skip a guardrail if you disagree, but you'll see it first. If /context/meta-patterns.md doesn't exist, skip this step silently.

Gate check — brain is missing or ICP is empty:

Do not block and redirect. Instead, run the Quick-Brain flow inline:

"I need three things before I can score beachhead candidates — this takes about 3 minutes. (Prefer the full 15-minute brain build instead? Run product-marketing-context and come back.)

1. Who's your best-fit customer today? One sentence — company type, size, and the pain that made them buy.
2. When they don't buy you, what do they do instead? Name the top 2–3 alternatives, including 'do nothing' if that's real.
3. What's the one thing you do that those alternatives can't?"

Wait for all three answers. Do not proceed to candidate identification until answered.

Write these three answers to /foundation/brain.md as a partial Section 2 (ICP) and Section 3 (Alternatives), tagged with a placeholder marker:

## Section 2: ICP Definition
🟡 Quick-Brain — built via beachhead-segment on [date]. Run `product-marketing-context`
to expand this into the full ICP profile (buying triggers, disqualifiers, willingness
to pay signals).

[user's answer to Q1]

## Section 3: Alternatives & Positioning
🟡 Quick-Brain — built via beachhead-segment on [date]. Run `product-marketing-context`
to expand into full competitive positioning.

Alternatives: [user's answer to Q2]
Differentiation: [user's answer to Q3]

Confirm before writing: "Here's what I'll save to your brain: [show exact text]. Save this?" Only write on explicit yes — same confirmation discipline as the full brain build (Operating Rule 1 in product-marketing-context).

Once written, proceed to Step 1 (Candidate Identification) using these answers.

Step 1 — Candidate Identification

Ask in one message. Never score before candidates are confirmed.

"Before I score, I need to understand the landscape:

1. What segments are you considering? Name 2–5 candidates. If not sure, describe your current customer base and I'll decompose it.

2. What evidence do you have? Named customers in each segment, win/loss patterns, inbound signals, sales feedback. Rough signals help.

3. What does winning look like in 12–18 months? Revenue target, reference customers, market share — be specific."

If the user names only one segment: challenge before proceeding.

"One segment isn't a choice — it's an assumption. Name at least two so we have something to compare. What else have you considered and rejected?"

Reflect back candidates and evidence:

"Scoring [N] segments: [list]. Evidence base: [summary]. Starting with Burning Pain."

Step 2 — Score Each Segment on Four Dimensions

Score each independently before comparing. Mark any dimension without evidence as [A] (assumption).

Dimension 1: Burning Pain (1–5)

Acute, urgent problem your product solves — or chronic inconvenience they've learned to live with?

Evidence required: named customer signal, sales call pattern, win/loss data, or inbound volume. If none: score ≤2, mark [A].

Dimension 2: Willingness to Pay (1–5)

Does this segment have budget, authority, and clear ROI justifying your price?

Cross-reference brain Section 5 proof points for ROI evidence. If missing: mark [A].

Dimension 3: Winnability (1–5)

Can you realistically reach 60%+ market share in this segment within 12–18 months?

Cross-reference brain Section 4 (Competitive landscape). If stale or empty: mark [A] — competitive positions cannot be inferred.

Dimension 4: Referral Potential (1–5)

Do customers in this segment talk to each other? Is one reference customer worth 10x their contract value in network effects?

This is what separates a beachhead from any other segment.

Step 3 — Apply Blocking Gates

Gate 1 — Pain floor (≥3): Any segment scoring <3 on Burning Pain is eliminated.

"❌ [Segment] eliminated — Burning Pain [X]. No urgency signal. GTM investment will be slow."

Gate 2 — Winnability floor (≥3): Any segment scoring <3 on Winnability is flagged as "Future Roadmap" — not current beachhead.

"⚠️ [Segment] flagged — Winnability [X]. You can win deals but cannot dominate. File as secondary. Revisit when competitive position improves."

Gate 3 — Assumption density: If >2 of 4 dimensions are marked [A] for top-scoring segment, confidence caps at 🟡.

"🟡 Conditional — [N] dimensions rely on assumptions. Validate [specific items] before committing."

Step 4 — Recommend Beachhead

After gates pass, state recommendation with rationale and expansion pathway.

## Recommended Beachhead: [Segment Name]
**Rationale:** [2–3 sentences grounded in four dimensions]
**Why not [Segment B]:** [One sentence specific reason]
**Why not [Segment C]:** [One sentence specific reason]

## What Winning Looks Like in 12–18 Months
- [Specific milestone 1 — revenue, customer count, or market share]
- [Specific milestone 2 — reference customer or analyst recognition]
- [Specific milestone 3 — expansion trigger]

## Expansion Pathway
**Beachhead → Adjacent 1:** [Segment] — [shared trigger or proof point needed]
**Adjacent 1 → Adjacent 2:** [Segment] — [what beachhead proves that unlocks this]

Trigger to move Beachhead → Adjacent 1: [specific milestone]
Trigger to move Adjacent 1 → Adjacent 2: [specific milestone]

## 90-Day Activation Plan
| Week | Action | Owner | Signal to watch |
|---|---|---|---|
| 1–2 | [Positioning action] | PMM | — |
| 3–4 | [First outbound step] | Sales/PMM | Meetings booked |
| 5–8 | [Pipeline build — target N qualified conversations] | Sales | Win rate |
| 9–12 | [First reference customer] | PMM | Inbound from segment |

## Eliminated Segments
| Segment | Gate | Reason |
|---|---|---|
| [X] | Pain floor | Burning Pain [score]: no urgency |
| [Y] | Winnability | Incumbent [name] entrenched |

Step 5 — Update Brain Section 2 (on Confirmation)

After recommendation confirmed:

"Updating brain Section 2 with [Segment] as confirmed beachhead. Confirm? [Y/N]"

Write to /foundation/brain.md Section 2:

## Beachhead Segment (last updated: [date])
- **Segment:** [name]
- **Scores:** Pain [X] / WTP [X] / Winnability [X] / Referral [X] — Total [X/20]
- **Confidence:** 🟢 / 🟡 / 🔴
- **Assumption flags:** [list or none]
- **Expansion pathway:** [Beachhead] → [Adjacent 1] → [Adjacent 2]
- **90-day trigger:** [first milestone]
- **Status:** Active beachhead

Never write without this explicit confirmation.

Step 6 — Learning Close

End every completed session by appending one row to /context/skill-sessions.md (create the file with a header row if it doesn't exist yet):

skill: beachhead-segment
session_date: [YYYY-MM-DD]
pattern: [one falsifiable statement about what happened this session, or "none"]
source: [surprised / wrong / missing / n.v.t.]

Write this row directly — do not ask the user for permission. This is a separate, mechanical log entry from the brain Section 2 write above, which still requires the user's explicit confirmation. If nothing notable happened this session, still write the row with pattern: none.

Outputs

• Files written: /foundation/brain.md Section 2 — the confirmed beachhead, scores, and expansion pathway (Step 5), only after explicit confirmation. /context/skill-sessions.md — one appended row per session (Step 6).
• Chat output format: Scorecard across four dimensions → blocking gate results → recommendation with expansion pathway, 90-day plan, and eliminated-segment reasons (Step 4 template).
• External side effects: None beyond the brain write and session log above.
• Next skill: After confirmed recommendation is written to brain, check product-marketing-context/.claude-plugin/next-skill-map.md for "After beachhead-segment" and surface that prompt. Do not auto-run the next skill — surface it as a question, wait for the user to say yes.

Verification

• Guardrails checked at Step 0 if /context/meta-patterns.md exists.
• At least 2 candidate segments scored — never a single segment in isolation.
• All 4 dimensions scored for every candidate, with [A] marking any dimension lacking evidence.
• Both blocking gates applied: Pain floor and Winnability floor.
• Every eliminated segment has a specific reason, not "didn't score well."
• Brain Section 2 write shown to the user and confirmed before it happens (Step 5).
• Session logged to /context/skill-sessions.md (Step 6).

Commands

/score [segment A] vs [segment B] vs [segment C]

Score named segments immediately. Skips intake.

/score fintech vs healthcare vs logistics

Output: Scorecard → gates → recommendation.

/decompose

Decompose current ICP into 2–4 scoreable sub-segments.

/decompose

/audit [segment name]

Pressure-test existing beachhead. Compares original vs. current scores. Surfaces expansion trigger status.

/audit

/pathway

Show full expansion pathway from confirmed beachhead.

/pathway

/eliminate [segment name]

Explicitly eliminate a segment. Documents reason.

/eliminate enterprise — too entrenched

Operating Rules

• Both Pain and Winnability must pass. No <3 on either becomes beachhead.
• Evidence over assertion. Dimension without signal = marked [A] and confidence reduced.
• Eliminated segments documented. Every elimination has specific reason, not "didn't score well."
• Expansion pathway mandatory. Beachhead is first step. Recommendation without clear stage 2 and 3 is incomplete.
• Never score one segment in isolation. Always require ≥2 candidates for comparison.
• Brain Section 2 write requires explicit confirmation. Never update silently.
• Challenge generic ICPs. "Mid-market B2B SaaS" is not scoreable. Require vertical, use case, or buyer role specificity.
• 90-day plan must have specific trigger. Name the number, behaviour, or event. "Build pipeline" is not a milestone.
• Referral Potential is the beachhead multiplier. Segment with 4+ Referral but moderate others is often still the right choice. Explicitly surface this trade-off.

Quality Gate

Do Not Use For

• go-to-market-strategy — after beachhead confirmed, use this to build full GTM brief
• positioning-messaging — for messaging for confirmed beachhead (run after)
• workflow-orchestrator — to chain this with above for full market entry program