| name | seo-content-brief |
| description | Generates a complete SEO content brief from a target keyword and SERP context. Includes search intent, AI Overview likelihood, outline with H2/H3 questions, internal/external linking, on-page metadata (title/meta/schema), and a 2026-aware angle. Use when the user says "write me a brief for keyword X", "I'm targeting [keyword]", or "outline a post about [topic]" and they want SEO traffic. |
/seo-content-brief
Produce a content brief that gives a writer everything they need to draft an SEO-ranked piece in 2026 — accounting for AI Overviews, E-E-A-T, topical authority, and the queries Google actually shows.
Usage
/seo-content-brief "best CRM for solo founders"
/seo-content-brief "cold email deliverability 2026" https://example.com/competitor-article
/seo-content-brief # interview first
If invoked with no arguments, ask before generating:
- Target keyword (primary + 2–3 secondary)?
- Search intent estimate: informational / commercial / transactional / navigational? (If unsure: I'll classify from the SERP — provide the SERP URL or paste top-3 results)
- Vendor's domain (so internal-link suggestions are real, not placeholder)
- ICP / target reader (who the article is FOR, not just "anyone who searches")
- Existing cluster pages? (helps internal-link suggestions and topical-authority strategy)
What this skill produces
A single markdown file containing:
- Intent classification — informational / commercial / transactional / navigational + supporting evidence
- AIO likelihood — high / medium / low + how to write for AIO citation (or avoid it)
- SERP analysis — what's currently ranking, why, what's missing
- The angle — what your piece does differently (no commodity content in 2026)
- Outline — H1, H2s as questions, H3s as sub-questions, word-count target per section
- Internal links — 5–10 suggestions (or placeholders for the vendor to fill)
- External links — 3–5 authoritative sources (the writer should fact-check + verify)
- On-page metadata — title tag (≤60 char), meta description (≤155 char), URL slug
- Schema markup — appropriate types (Article + FAQ + HowTo / Product / Review)
- Promotion plan — how the piece gets links, mentions, social pickup (without promo, no ranking)
Intent classification (the foundational decision)
You can't write to rank without knowing intent. The four:
- Informational — "what is X", "how does X work", "X explained". User wants to learn.
- Commercial — "best X", "X vs Y", "X for Z use case", "X reviews". User is researching to buy.
- Transactional — "buy X", "X pricing", "X coupon", "X login". User is ready to act.
- Navigational — "X.com", "X company", branded. User knows where they're going.
Mixing intent is the #1 SEO writing mistake. A "best CRM" article (commercial) that opens with 800 words of "what is a CRM" (informational) will not rank in 2026 — Google rewards intent alignment, AIO kills the front-loaded informational fluff.
To classify:
- Look at top 10 organic results (you'll see this in the SERP analysis section)
- Match patterns: 10 listicles = commercial; 10 long-form explainers = informational; mix = mixed intent (rare, hard)
- Verify with PAA (People Also Ask) — what does Google show as sibling questions?
AIO likelihood (the 2026 wrinkle)
AI Overviews appear on ~30–50% of US informational queries (varies by vertical). Strategy differs:
- High AIO likelihood (informational, generic):
- Re-evaluate: is this query worth chasing? Click-through is down 30–60%.
- If yes: write to BE the citation source. Lead with a TL;DR. Use clear H2-as-question structure. Use original data.
- Goal: appear inside the AIO answer (branded citation), not below it (where clicks are dead).
- Medium AIO likelihood (some informational mix):
- Standard SEO best practices apply.
- Add original angle, E-E-A-T signals, comparison tables.
- Low AIO likelihood (commercial, transactional, navigational):
- AIO usually doesn't render. Standard SEO works.
- Focus on comparison + persona + use-case angles. These are AIO-resistant.
You estimate AIO likelihood by:
- Search the actual SERP — does AIO render? (this is the source of truth)
- Or: estimate from query type. "What is X" = ~80% AIO probability. "Best X for Y" = ~20%.
SERP analysis (do this before outlining)
Read the top 3–5 organic results (the user provides URLs, OR you note "I need top-3 URLs to do this properly"). For each:
- Word count
- Author / publication credibility (E-E-A-T signal)
- Date of last update (Google rewards freshness in many verticals)
- Format (listicle, long-form, comparison, tutorial)
- Unique angle / data / original sources
- What they're MISSING (this is where your piece wins)
If you don't have SERP access (no WebFetch / user didn't provide URLs), say so and tell the user to paste the top 3 URLs OR a SERP screenshot. Don't guess what's ranking.
The angle (the part that gets you cited / ranked)
In 2026, the question is not "can I write a comprehensive piece about X" — it's "what unique angle do I bring that the top 5 results don't already have."
Common angles that work:
- Original data — survey of 200 customers, scrape of 1,000 listings, internal analytics
- First-person experience — "I did X for 90 days, here's what happened"
- Contrarian take — "Everyone says X. We did Y and got better results"
- Updated post-event — "What changed after Google's Feb 2024 update"
- Specific persona/use-case — "Best CRM for solo agency owners" (vs generic "best CRM")
- Comparison the SERP is missing — "X vs Y vs Z" where 2 of those don't currently rank against each other
If the brief doesn't have an angle, the writer will produce a commodity piece. Always specify the angle, never leave it as "find an angle when you write."
Outline structure (the 2026 standard)
H1: [Title — matches title tag, ≤ 60 chars]
[TL;DR — 60–80 words, answers the literal query]
[Why we wrote this — 1 sentence, establishes E-E-A-T author/experience]
H2: What is [primary keyword]? (if informational mix, 60–100 words MAX — give AIO what it needs and move on)
H2: [First commercial question — e.g., "When should you use [X]?"]
H3: [Sub-question 1]
H3: [Sub-question 2]
- List or comparison table here
- 200–400 words this section
H2: [Second commercial question — e.g., "Top [N] options compared"]
- Comparison table (great for AIO citation)
- 100–150 words intro
- Then table with explicit columns
[ continues with intent-aligned H2s ]
H2: How to choose / What to look for / How to set up [X]
- Concrete steps or framework
- Goal: be the practical answer
H2: Common mistakes / What goes wrong
- Specific, contrarian, with examples
- Drives engagement signals (saves, shares)
H2: FAQs (schema-marked)
- 4–8 questions (mine from PAA + Reddit + customer support tickets)
- 40–80 words per answer (snippet-ready)
[Author bio — name, photo, credentials, links to LinkedIn — E-E-A-T essential]
[Last updated date]
Per-H2 word-count guidance keeps writers honest. Without them, sections balloon and lose intent alignment.
Internal and external links
Internal: link to 5–10 of the vendor's own pages, biased toward:
- Pillar page for this cluster (always)
- 2–3 cluster pages that drill into sub-topics
- 1–2 commercial pages (pricing, product, demo) — but NOT in the first 30% of the article
- 1–2 case studies that demonstrate experience
If the vendor has no existing pages, mark internal-link suggestions as [NEEDS: cluster page on subtopic X] so they know what to write next for topical authority.
External: link to 3–5 authoritative sources:
- Original research (Gartner, Forrester, McKinsey reports — only the freely-citable parts)
- Government / standards docs (RFC, ISO, FTC)
- Authoritative practitioner blogs (well-known names in the space)
- AVOID linking to direct competitors unless absolutely necessary
External links signal that you're not a closed-loop content farm. Pages with 3–5 outbound links rank better than 0-outbound pages.
On-page metadata
Title tag (60 chars max):
- Primary keyword in first 30 chars
- Brand at the end (optional, only if recognized)
- Number / specificity / promise
Examples:
- ✅ "Best CRM for Solo Founders (2026): 7 Real Comparisons"
- ✅ "Cold Email Deliverability After Google's 2024 Update"
- ❌ "The Complete Guide to CRM" (generic)
Meta description (155 chars max):
- Restate the primary keyword
- Promise the value
- End with implicit CTA
URL slug:
- 3–5 words, lowercase, hyphens
- Primary keyword + qualifier
- Examples:
/best-crm-solo-founders/, /cold-email-deliverability-2024/
Canonical: self-referential unless multiple URLs exist
OG title / OG description: match meta but slightly more "social"
Schema markup
Always: Article schema with author, publisher, datePublished, dateModified.
Conditionally:
- If FAQ section:
FAQPage schema
- If step-by-step:
HowTo schema
- If comparing products:
Product + Review + AggregateRating (only if you actually have ratings, don't fake)
- If listicle of products:
ItemList
Provide schema JSON-LD snippets in the brief (the writer/developer drops them in).
Promotion plan (without this, the piece sits)
The brief should include a 5-item promotion plan:
- Internal: link to it from 3–5 existing high-authority pages on the vendor's site
- Newsletter: feature in next send
- Social: 2 LinkedIn posts + 1 X thread + 1 carousel in next 14 days
- Earned: pitch as a guest post or "data source" to 5 publications (digital PR)
- Community: post in 1–2 relevant subreddits (only if you've earned karma there) OR 1 Hacker News if relevant
A piece with zero promo sits at position 25–40 indefinitely. A piece with 3+ external mentions in the first 30 days has a real shot at top 10.
Output format
Write to seo-brief-{slug}.md:
# SEO Brief: {Title}
**Primary keyword**: cold email deliverability 2026
**Secondary keywords**: DMARC outbound 2024, cold email warmup, Google Yahoo bulk sender
**Search volume (est.)**: 1.2K–2.4K/mo (cite tool: Ahrefs / SEMrush)
**Keyword difficulty**: 32/100 (medium)
**Intent**: Mixed informational + commercial
**AIO likelihood**: Medium — appears on ~40% of related queries
## SERP Analysis
Top 3 current rankers:
1. Postmark blog — "Email deliverability guide" — 2,800 words, 2023 last updated, author = engineering. Strong but stale on Feb 2024 update.
2. Mailchimp — "Email deliverability best practices" — 1,800 words, generic, no original data.
3. Litmus — "DMARC 101" — 1,500 words, technical but doesn't cover cold outbound specifically.
Gap: nothing top-ranking specifically addresses cold-outbound deliverability post-Feb-2024 with original sender data.
## The Angle
Original data: post-mortem of [Vendor's own] cold outbound recovery in March–April 2024.
Numbers: reply rates went 9% → 2% → 6% over 90 days.
Specific actions that mattered + ones that didn't.
This is what nothing in the top 5 currently has.
## Outline
H1: Cold Email Deliverability After Google's Feb 2024 Update (What Actually Changed)
TL;DR (80 words):
> In Feb 2024 Google and Yahoo enforced DMARC + one-click unsubscribe + sub-0.3% spam rate for bulk senders.
> Cold outbound reply rates dropped 40–70% across the industry. We did the post-mortem on [Vendor]'s own
> ramp-back: which fixes worked, which didn't, the 90-day recovery curve. If your cold email opens cratered
> in spring 2024, this is the playbook. (5-min read.)
[ ... 8 H2 sections, each with H3 sub-questions and 200–600 word target ... ]
## Internal Links (5)
- [Pillar] /b2b-outbound-playbook/ (link from H2 #2)
- [Cluster] /dkim-rotation-guide/ (link from H2 #3)
- [Case study] /how-we-recovered-9-percent-reply-rate/ (link from H2 #4)
- [Product/Demo] /book-a-call/ (link from final CTA only)
- [NEEDS — write next]: /list-hygiene-for-cold-outbound/ (link from H2 #5)
## External Links (4)
- Google Email Sender Guidelines (official)
- Yahoo Sender Best Practices (official)
- DMARC.org Deployment Guide
- Postmark blog: transactional separation pattern
## Metadata
- **Title tag**: Cold Email Deliverability 2026: What Changed (Recovery Playbook)
- **Meta description**: Google's 2024 update broke cold outbound. Here's what we did to recover reply rates, with 90-day data and the specific DMARC/list-hygiene checklist.
- **URL slug**: /cold-email-deliverability-2024/
- **OG image**: Custom (data chart of reply rate recovery)
## Schema
```json
{
"@context": "https://schema.org",
"@type": "Article",
"headline": "...",
"author": { ... E-E-A-T author bio ... },
...
}
(Plus FAQPage schema for the FAQ section.)
Promotion Plan
- Internal: link from /b2b-outbound-playbook/ within 24h of publish
- Newsletter: lead piece in next Friday's send
- LinkedIn: 2 text posts + 1 carousel over 7 days
- X: 1 long-form thread Day 1, 3 atomic posts over 10 days
- Outreach: pitch to Smartlead / Lemlist / Instantly blogs as guest data piece
Estimated production cost
- Writer: 6–10 hours
- Editor: 1–2 hours
- Custom chart: 1 hour
- Total: ~$400–800 if outsourced, ~half day if internal
## Workflow
1. **Confirm the keyword + intent + vendor domain + ICP**. If missing, ASK.
2. **Pull SERP context**:
- User-provided URLs (preferred — fetch and analyze)
- If WebFetch available, fetch top 3
- If neither: tell the user to paste top-3 URLs, then continue
3. **Search the KG**:
- `hybrid_search("SEO AI Overviews E-E-A-T")` — get the 2026 rules
- `hybrid_search("copywriting frameworks")` — apply to article structure
- `hybrid_search("ICP buyer persona")` — make sure the brief matches the audience
4. **Classify intent** from the SERP. Don't skip this.
5. **Estimate AIO likelihood**. Adjust strategy accordingly.
6. **Find the angle**. If you can't find one in 5 minutes of thinking, ask the user "what's unique about how YOU approach this?" — the angle usually comes from their experience.
7. **Build the outline** — H2 questions, H3 sub-questions, word-count targets.
8. **Generate metadata + schema + internal/external links + promo plan**.
9. **Write the file** — one markdown file at the path the user specified (default: `seo-brief-{slug}.md`).
10. **Report back** — file path + the angle + the one external link the writer must verify before publishing. Don't dump the brief into your reply.
## What this skill is NOT
- Not a writer. You produce the brief; the user/writer produces the article.
- Not a keyword researcher. The user provides the target keyword (or asks `/keyword-research` separately — not yet built).
- Not a Google. You don't have live SERP access unless WebFetch is available and the user authorizes URLs.
- Not a backlink-build tool. You suggest a promo plan; the user executes.
## Common mistakes
- ❌ Skipping intent classification (writes the wrong piece)
- ❌ Ignoring AIO likelihood (writes top-funnel content that's eaten by AIO)
- ❌ No unique angle (commodity piece, won't rank)
- ❌ Generic metadata ("Complete Guide to X")
- ❌ Internal links to nonexistent pages without marking `[NEEDS: ...]`
- ❌ Outline with H2s as statements instead of questions (poor AIO citation, poor PAA capture)
- ❌ No FAQ section (misses easy schema win)
- ❌ Forgetting the promo plan (piece without promo doesn't rank)
## Knowledge graph access
Search before drafting:
- `hybrid_search("SEO 2026 AI Overviews E-E-A-T")` — the 2026 rules
- `hybrid_search("copywriting frameworks")` — for the section-by-section persuasion
- `hybrid_search("ICP buyer persona")` — to keep the brief audience-specific
- `hybrid_search("social platform algorithms")` — for the promo plan
## Success criteria
You succeed when:
- The writer reads the brief and starts drafting without asking a single clarifying question
- The piece has a clear angle (not commodity)
- Intent is correctly classified
- AIO strategy is explicit (write FOR the citation OR avoid the keyword)
- Metadata fits the character limits
- Internal links are real (or marked NEEDS for the topical authority gap)
- The promo plan exists and is actionable
