Plans SEO as a GTM acquisition motion — keyword strategy, content architecture, and distribution plan grounded in ICP demand signals. Produces a content strategy that compounds over time rather than single-post tactics. Use when organic search is a viable acquisition channel for the ICP and team has capacity for sustained content. NOT a technical SEO audit.
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Une commande directe contourne le prompt de vérification. Examinez la source avant de l'exécuter.
Plans SEO as a GTM acquisition motion — keyword strategy, content architecture, and distribution plan grounded in ICP demand signals. Produces a content strategy that compounds over time rather than single-post tactics. Use when organic search is a viable acquisition channel for the ICP and team has capacity for sustained content. NOT a technical SEO audit.
triggers
["/seo-content-strategy","user asks about SEO strategy, content strategy, or organic growth","team is evaluating SEO as a growth channel","content is being produced but not driving search traffic or signups","user asks which keywords or topics to target"]
role
workflow
version
1.0.0
sources
["SEO as GTM motion synthesis (research synthesis, 2026-06-13)","Jobs-to-be-Done (Christensen) — demand state framing for keyword strategy"]
Role: Organic growth architect. You connect what your ICP is searching for to what your product does — then build a content architecture that captures that demand systematically. Not a checklist of "write more blog posts." A motion with intent.
Before starting
Confirm (ask or infer):
ICP card available? Specifically: Layer A (JTBD), Layer C (alternatives), Layer D (deciding language). These define the demand language that becomes keyword strategy.
Product stage — SEO compounds over 6–18 months. Is the team committing to sustained production, or looking for a one-quarter initiative?
Existing domain authority / content — is there a current site with pages indexed? If domain authority < 10 and team cannot produce content for 6+ months, SEO is likely not the right primary acquisition motion at this stage.
Technical infrastructure available? — CMS, blog, developer docs. Not evaluating technical SEO here, but content needs a home.
IF product_stage = pre-pmf OR icp_not_defined →
BLOCK. Return:
"SEO strategy requires a defined ICP with validated JTBD (Layer A) and deciding
language (Layer D). Without these, keyword strategy is speculation.
Run /icp-research first."
IF team_content_capacity = 0 OR committed_months < 6 →
WARN. Return:
"SEO compounds slowly — median time to ranking for new content is 4–12 months.
If team cannot sustain content production for 6+ months, SEO is a poor
primary acquisition channel at this stage. Consider: paid (see /paid-channel-fit),
community, or developer sponsorships for faster signal."
Inputs
Input
Required?
Description
ICP card (Layer A + D)
Required
JTBD and deciding language → demand language
ICP alternatives (Layer C)
Required
Competitor search terms; category vocabulary
Existing domain + content
Required
Baseline authority and gaps
Content capacity
Required
Hours/week or pieces/month sustainable
Revenue stage
Recommended
Pre-PMF / Early growth / Scaling — determines keyword ambition
Contract
This skill guarantees:
Keyword strategy is grounded in ICP demand language, not keyword volume alone
Content architecture separates demand capture (existing search intent) from demand creation (no current search)
Topic clusters are organized to build authority in one space before expanding
Output includes a content velocity estimate: what can be produced at current capacity
All content recommendations require pmm/content-review before publishing
Decision logic
Phase 1 — Demand state mapping
Map ICP's search behavior to three demand states:
Demand State 1 — Problem-aware (searching for the pain):
What ICP searches when they know they have the problem but haven't found the solution.
Source: ICP Layer A (JTBD) and Layer D (deciding language).
Keyword pattern: "[pain] how to", "[problem] fix", "[situation] causes [symptom]"
Content type: Educational / diagnostic. High volume, competitive, slower to rank.
Examples from infrastructure ICP: "postgres slow query production", "database regression after deploy"
Demand State 2 — Solution-aware (comparing options):
What ICP searches when they know solutions exist and are evaluating.
Source: ICP Layer C (alternatives considered).
Keyword pattern: "[product] vs [competitor]", "[category] alternatives", "best [category] tool"
Content type: Comparison, alternative, vs. page. Lower volume, higher intent, faster to convert.
Demand State 3 — Category-defining (the ICP doesn't know to search):
Problems your product solves that the ICP doesn't have a name for yet.
Source: Deciding language from Layer D — phrases that are unique to your customers.
Content type: Thought leadership, vocabulary-setting. No search volume yet; builds category.
Note: DO NOT optimize this for current keywords; optimize for future demand capture.
Phase 2 — Keyword cluster architecture
Organize keywords into topic clusters with a pillar + spoke structure:
Pillar page (1 per cluster):
- High-volume, competitive head term for the primary pain
- Comprehensive; answers the full problem, not just one angle
- Links to spoke content
- Target: domain authority + 20% (what you can realistically rank for)
Spoke pages (5–10 per cluster):
- Long-tail variations and specific sub-problems
- Lower volume but faster to rank; high conversion intent
- Each links back to the pillar
RULE: Build one complete cluster before starting a second. Partial clusters rank worse than complete ones.
Cluster selection criteria:
1. Primary cluster: the cluster closest to the ICP's trigger event (highest conversion)
2. Secondary cluster: the ICP's most common alternative (comparison / vs. content)
3. Tertiary cluster: the ICP's JTBD vocabulary (Layer D phrases as search terms)
Start no more than 2 clusters simultaneously.
Phase 3 — Content velocity and prioritization
Prioritization rule (weekly budget → what to produce):
IF content_capacity ≤ 2 pieces/month →
Focus: spoke pages only (long-tail, fastest to rank).
Do NOT start pillar pages until capacity increases.
Rationale: Incomplete pillars rank poorly; consistent spoke production compounds faster.
IF content_capacity = 2–4 pieces/month →
Mix: 1 pillar + 3 spokes per cluster per quarter.
Sequence: research → pillar → spokes → internal linking pass.
IF content_capacity > 4 pieces/month →
Build 1 complete cluster per quarter; add comparison content once cluster is live.
Add: SEO-driven thought leadership (Demand State 3) when 1 cluster is ranking.
QUALITY CONSTRAINT:
One well-researched, high-quality piece ranks better than four thin pieces.
Do not produce content below 1,200 words for spoke pages or 3,000 words for pillar pages.
All content passes /content-review before publishing.
Phase 4 — Distribution amplification
SEO content compounds when distributed beyond organic search. After publishing:
Distribution plan per piece (adapt to available channels):
- Developer community posting: HN, relevant subreddits, Discord communities —
post on the problem, reference the article
- LinkedIn: excerpt for the ICP persona (not the headline — a specific insight from the piece)
- Internal linking: update existing content to link to the new piece
- Outbound use: high-value pieces become touch 4–5 in the outbound sequence (value-before-ask)
- Email: existing subscriber list or waitlist if applicable
RULE: Distribution effort should equal creation effort on the first week of publication.
An unread piece that ranks positions = wasted distribution leverage.
Output format
## SEO content strategy — [date]
ICP: [1-line segment from ICP card]
Content capacity: [pieces/month]
Domain authority estimate: [current / target]
### Demand state map
State 1 — Problem-aware keywords: [top 5–8]
State 2 — Solution-aware keywords: [top 5–8]
State 3 — Category-defining vocabulary: [top 3–5 phrases from Layer D]
### Cluster architecture
Primary cluster: [head term + 5 spokes]
Secondary cluster: [head term + 5 spokes] (start after primary is complete)
### Content calendar (next 3 months)
Month 1: [specific pieces — title, keyword target, demand state]
Month 2: [specific pieces]
Month 3: [specific pieces]
### Content velocity rule
[Which prioritization rule applies at current capacity]
### Distribution plan
[Per piece: which channels, what angle for each channel]
### Success metrics (check at 6 months)
Keyword positions: [specific target keywords]
Organic traffic: [baseline + target]
Content-attributed pipeline: [how to measure]
Cross-review triggers
Required before any content is published:
pmm/content-review/ — banned hype patterns, claims accuracy, brand voice on every piece
Anti-patterns
Anti-pattern
Why it fails
Fix
Keyword strategy based on volume alone
High-volume keywords are dominated by large domains; low-intent content attracts readers who never convert
Weight for conversion intent (Demand State 2) and ICP demand language (Layer D), not just volume
Starting 3 topic clusters simultaneously
Each cluster is incomplete; partial clusters rank poorly and authority doesn't compound
Build one complete cluster before starting a second
Publishing thin content quickly
Google's quality signals penalize thin content; it dilutes domain authority
1 high-quality piece ranks better than 4 thin ones; see content velocity rules
Creating content without distributing it
Organic search takes months; content needs to drive traffic on day 1 via distribution
Spend equal time on distribution as creation in the first week
Measuring success at 2 months
SEO ranking timeline is 4–12 months; early measurement produces false negatives
Set a 6-month first-checkpoint; track position movement, not just traffic
"We'll do SEO later when we have time"
SEO compounds — delaying by 6 months = losing 6 months of compounding
If organic is a strategic channel, start now at small capacity; do not wait for spare capacity
Validation criteria
All keyword clusters grounded in ICP JTBD (Layer A) or deciding language (Layer D) — not volume alone
Demand state map distinguishes problem-aware, solution-aware, and category-defining keywords
Content velocity rule matches current team capacity — no over-ambitious calendar
Cluster architecture: one primary cluster defined end-to-end before secondary
Distribution plan present for at least the first month of content
Success metrics defined with 6-month checkpoint
Benchmarks (2025–2026)
Benchmark
Value
Source
Median time to rank for new content (pages 1–10)
4–12 months
Ahrefs research 2025
Topic cluster approach vs. individual posts (traffic)
2–3× more organic traffic
HubSpot topic cluster research
Comparison / vs. content conversion rate vs. educational
3–5× higher
Clearbit / Segment research
Content pieces needed for SEO to become primary channel
