design-citation-magnet-content

# Design Citation-Magnet Content This skill produces a content brief for a piece of content designed to win citation in AI-mediated search. The brief is more demanding than a standard content brief — it specifies primary-evidence requirements, comparative structure, source citations, schema, and per-engine targeting in addition to the usual outline / word count / internal-link spec. The skill is opinionated: the brief requires a green-lit topic from `plan-generative-seo-strategy` (or a documented exemption); primary evidence (original photography / testing / surveys / data) is required where the strategic plan claimed the moat — without it, the brief defaults to generic explainer territory; sources are named, not gestured at; schema is specified; brand voice holds. ## When to use this skill - A topic has been picked in `plan-generative-seo-strategy` and is ready to brief. - The user wants a content brief specifically for AI-citation, not for generic ranking. - The user has primary-evidence assets (or is prepared to produce them) for the topic. - The user wants to bake citation patterns into a sibling content brief (the standard brief in Content SEO leaf 08 is a less-demanding template; this one is for citation-priority topics). ## When NOT to use this skill - No `plan-generative-seo-strategy` output exists for the cluster — run that first. Briefing without strategy wastes budget. - The topic was on the strategy's "skip" list — don't brief topics the planner explicitly rejected. If the user wants to override, surface the disagreement and require a documented rationale. - The user wants a generic content brief — use `generate-content-brief` in Content SEO leaf 08. - The user wants to audit / rewrite existing content — use `audit-ai-content-readiness`. - The user has no capacity for primary evidence and the topic's defensibility argument requires it — flag the constraint; recommend either acquiring evidence first or picking a different topic. ## Inputs required 1. **Business context and SEO foundation** — read `businesses/<slug>/business_context.md` and, for SEO-specific work, `businesses/<slug>/seo_foundation.md`. Resolve `<slug>` by listing `businesses/`: if exactly one folder exists, use it; if multiple, ask which business; if none, recommend running `business-context` first. If SEO foundation is missing, recommend running `seo-foundation` before producing output. Sections this skill cares about most: 3 (products), 4 (customer + questions), 6 (competitors), 8 (brand voice — non-negotiable for this skill). 2. **Strategic plan** — required. Path to the relevant `ai-seo/generative-seo-strategy-*.md`. The brief consumes the plan's defensibility argument, per-engine targets, and investment estimate. If absent, refuse and recommend running `plan-generative-seo-strategy` first. 3. **Topic** — required. Which pick from the strategic plan is being briefed? Cross-reference the plan's win-able picks; reject if the topic is on the skip list. 4. **Priority queries** — required. The 2-6 queries this content targets. From the strategic plan. 5. **Primary-evidence inventory** — required if the strategic plan's defensibility argument depends on primary evidence. What's available, what needs to be produced, what the production timeline looks like. 6. **Cluster + internal-link context** — required. Where in the cluster does this content sit (pillar / spoke / standalone)? What pages should it link to and from? Cross-reference `clusters/cluster-map.md` if it exists. 7. **SERP + competitor citation reference** — required. From the strategic plan; what are the currently-cited sources, and what's the brief's plan to displace or join them? If the strategic plan is missing or the topic is on its skip list, refuse with a clear explanation and recommend the appropriate alternative. ## Process 1. **Verify the topic against the strategic plan.** - Is this topic on the strategic plan's win-able list? - If marginal: surface the marginality risk in the brief and proceed. - If skip-listed: refuse; the strategic plan said don't compete here. 2. **Frame the question.** - State the topic as the question a user would actually ask, not as a topic label. - Cross-reference the priority query phrasing — the H1 and H2s should match query phrasing. 3. **Design the answer block.** - **Definitive opening sentence:** the first sentence after the H1, answering the question directly. Specific, sourced, on-voice. Draft the actual sentence. - **TL;DR:** 3-5 bullet summary block immediately under the answer sentence. Bullets cover: the headline answer, the most-important nuance, the comparison context, the brand-evidence note. 4. **Plan the structural body.** - Question-led H2s. List 6-10 specific H2s the article should cover. Each H2 framed as a user question. - Per H2: a one-line direction, the key entities to name, the primary-evidence asset that supports the section (if applicable), the source citation list. 5. **Specify primary evidence.** - What original photography / testing / data / surveys is required? - What does production look like (photographer? Customer recruitment? Studio time? Survey instrument?)? - What's the timeline? - What does the asset look like in-page (table? Image grid? Embed? Downloadable?)? - This is the moat — under-specifying here wastes the budget. 6. **Specify comparative structure.** - Most citation-magnet content benefits from a comparative table or grid. - List the comparison axes (e.g. for sunscreen: SPF level, active ingredient, concentration, tint range, white-cast outcome per Fitzpatrick type, price-per-mL, certifications). - List the products / sources being compared (own product + 2-4 competitors). 7. **Specify source citations.** - Name the specific authoritative sources to cite (FDA OTC Sunscreen Monograph; Skin Cancer Foundation; specific peer-reviewed studies — author + year + journal; established publications). - Don't gesture at "studies show" — name the studies. - 6-15 citations typical for a depth piece. 8. **Specify schema.** - `Article` with explicit `about` and `mentions` properties. - `ImageObject` per primary-evidence photograph with `caption`, `creator`, and `contentUrl`. - `FAQPage` *only* if the article contains a genuine FAQ section. - `Product` `mentions` if the article references named products (own + competitor). - `BreadcrumbList` for cluster context. 9. **Specify internal-linking plan.** - From this article: 3-6 outbound internal links with descriptive anchors (to PDPs, cluster pillar, related spokes). - Inbound: which existing pages should link to this article, with what anchor text. - This is part of the entity-graph signal — under-spec here weakens the citation eligibility. 10. **Specify freshness signals.** - Visible "Last updated: [planned publish date]" line. - For time-sensitive topics, a one-line update-reason note (e.g. "Updated 2026-05 with new FDA OTC monograph guidance and original photography across Fitzpatrick I-VI skin tones"). - Quarterly review schedule baked into the brief. 11. **Specify per-engine targeting.** - Per priority query: which engines (AI Overviews / Perplexity / ChatGPT / Claude) we're targeting. - Currently cited sources we're trying to displace or join. - Citation-pattern notes (e.g. "Perplexity cites generously — 5+ citations means high probability of inclusion if structure is clean; ChatGPT cites sparingly — focus on definitive opening + sourced claims"). 12. **Brand voice constraints.** - Quote the relevant brand voice voice rules from the profile. - Field & Sun: ingredient-led, plain-language, specific. Reject vague category language, unsupported safety claims, and inflated promise language. 13. **Quality gates + reviewer checklist.** - Pre-publish gates: legal review for claims; brand-voice review; technical-SEO review for schema; primary-evidence consent / rights confirmation. 14. **Production timeline + capacity check.** - Match the strategic plan's investment estimate. - Specify which week each component (photography / writing / design / schema / internal-link / promotion) lands. 15. **Write the brief.** ## Output format ```markdown # Citation-Magnet Brief — [Topic name] **Brief date:** [date] **Cluster:** [cluster name]. **Strategic plan:** [path to `ai-seo/generative-seo-strategy-*.md`]. **Pick reference:** [Pick N from the strategic plan]. **Author:** [name / role]. **Target publish date:** [date]. ## Topic + question framing **Topic:** [topic name]. **User question (H1):** [actual question phrasing]. **Priority queries this content targets:** - [query 1] (vol [n]/mo) - [query 2] (vol [n]/mo) - [query 3] (vol [n]/mo) - [query 4] (vol [n]/mo) ## Defensibility argument (from strategic plan) [1-2 paragraphs from the plan: what does the brand have that competitors don't? What's the moat? What's the citation strategy — displace, join, or expand the cited-sources set?] ## Per-engine targeting | Query | AI Overview | Perplexity | ChatGPT | Claude | Currently cited (to displace / join) | |-------|-------------|------------|---------|--------|--------------------------------------| | ... | ... | ... | ... | ... | ... | ## Answer block (draft) **Definitive opening sentence:** > [Draft the actual sentence. Specific, sourced, on-voice. ~25-40 words.] **TL;DR:** - [Bullet 1: headline answer.] - [Bullet 2: most-important nuance.] - [Bullet 3: comparison context.] - [Bullet 4: brand-evidence note (if applicable).] ## Structural body — H2 outline ### H2 1: [Question phrasing] - **Direction:** [one-line on what the section establishes]. - **Entities to name:** [specific entities, with concentrations / qualifiers where relevant]. - **Primary evidence:** [photo / table / data point]. - **Source citations:** [named sources with link or full reference]. ### H2 2: [Question phrasing] [Same structure] [6-10 H2s total] ## Primary-evidence specification **Asset 1: Original photography across Fitzpatrick I-VI skin tones** - **Production:** photographer [N] week(s); 8 customer participants recruited; consistent lighting methodology. - **Output:** 24 photographs (8 customers × 3 product applications); side-by-side comparison grid. - **Timeline:** weeks 1-2 of production. - **Rights:** written consent for editorial use from each customer. - **In-page treatment:** comparison grid with caption per photo specifying customer Fitzpatrick type, product applied, lighting condition. **Asset 2: Comparative table** - **Axes:** SPF level | Active ingredient | Concentration | Tint range | White-cast outcome (Fitzpatrick I-VI) | Price-per-mL | Certifications - **Products compared:** Field & Sun Mineral Sun Drops SPF 50; Supergoop Mineral Mattescreen; ILIA Super Serum SPF 40; Saie Sunvisor SPF 35. - **Source for competitor data:** product packaging / brand site / certification database (cited inline). [Repeat per asset] ## Comparative structure [Specify the table or grid structure that the body will use. Most citation-magnet content benefits from at least one comparative element — extract-friendly for AI systems.] ## Source citations (named) | # | Source | Type | Use in article | |---|--------|------|----------------| | 1 | FDA OTC Sunscreen Monograph (2024) | Regulator | Active-ingredient classification + concentration limits | | 2 | [Author et al, 2023, Journal] — "Iron oxide tinting in mineral SPF formulations" | Peer-reviewed study | Tinting-mechanism section | | 3 | Skin Cancer Foundation — "Mineral vs Chemical Sunscreen Guide" | Authority | Mechanism context | | ... | ... | ... | ... | ## Schema specification - **`Article`:** `about: "Tinted Mineral Sunscreen for Darker Skin Tones"`; `mentions: [Mineral sunscreen, iron oxide, zinc oxide, Fitzpatrick scale, Daily Glow Serum, Mineral Sun Drops SPF 50]`; `dateModified: [planned publish date]`. - **`ImageObject` per photograph:** `caption`, `creator`, `contentUrl`, `representativeOfPage` for hero image. - **`Product` `mentions`:** named products in the comparative table. - **`BreadcrumbList`:** Home > Sun Care > [Cluster pillar] > [This article]. - **No `FAQPage`** unless a genuine FAQ section is included; do not manufacture Q&A for schema eligibility. ## Internal-linking plan **Outbound links (from this article):** | # | Anchor | Target | |---|--------|--------| | 1 | "Mineral Sun Drops SPF 50" | /products/mineral-sun-drops-spf-50 | | 2 | "no white cast guide" (cluster pillar) | /blog/mineral-vs-chemical-sunscreen | | 3 | "iron oxide tinting in skincare" (related spoke) | /blog/iron-oxide-tinting-explainer | | ... | ... | ... | **Inbound links (to update):** | # | From page | Anchor | |---|-----------|--------| | 1 | Cluster pillar | "tinted mineral sunscreen for darker skin tones" | | 2 | Sun Care collection page | "no white cast on darker skin tones" | | ... | ... | ... | ## Freshness specification - **Visible last-updated line:** "Last updated: [date]" displayed under H1. - **Update-reason note:** "Updated 2026-05 with original photography across Fitzpatrick I-VI skin tones and current FDA OTC monograph guidance." - **Review cadence:** quarterly (next: Q3 2026). ## Brand voice constraints (non-negotiable) - Ingredient-led, plain-language, specific. Reject vague category language, unsupported safety claims, and inflated promise language. - Use specific concentrations (15% L-ascorbic acid; 20% non-nano zinc oxide) rather than gestural ingredient claims. - Tone: warm, knowledgeable, plain-language. Confident without being preachy. ## Quality gates / reviewer checklist - [ ] Definitive opening sentence drafted and approved. - [ ] TL;DR drafted and approved. - [ ] All H2s framed as user questions. - [ ] Primary-evidence assets produced and consent on file. - [ ] All factual claims have named source citations. - [ ] Schema spec implemented and validated (no schema-content mismatch). - [ ] Internal links wired in both directions. - [ ] Last-updated date and update-reason note rendered visibly. - [ ] Brand voice review passed (no vague category-language or unsupported-claims residue). - [ ] Legal review passed (FDA OTC drug claims; customer photography rights). - [ ] Technical SEO review passed (schema validation; canonical alignment; mobile rendering). - [ ] Pre-publish QA: read on mobile and desktop; verify TL;DR renders cleanly; verify comparative table is responsive. ## Production timeline | Week | Activity | Owner | |------|----------|-------| | 1 | Photographer recruitment + customer recruitment | Marketing lead | | 2 | Photography shoot + post-production | Photographer | | 3 | First-draft writing | Writer | | 4 | Comparative-table research + data validation | Writer + founder | | 5 | Editorial review + brand-voice review + legal review | Marketing lead + founder | | 6 | Schema implementation + internal-linking + technical SEO review | Dev (contracted) | | 6 | Publish + GSC submission + promotion brief | Marketing lead | ## Success metrics + measurement plan - **At 30 days post-publish:** confirm indexing; track ranking on priority queries. - **At 60 days post-publish:** re-run `audit-llm-visibility` for the priority queries; capture per-engine citation rate. - **At 90 days post-publish:** evaluate against strategic plan's expected lift; decide whether to invest in adjacent topics in the cluster. ## Risks + watchouts 3-5 specific risks. E.g. "Customer photography consent must be in writing per customer; legal review pre-publish — without consent, the moat asset cannot ship"; "Photographer / production capacity is on the critical path; build a 1-week buffer"; "Schema must validate cleanly — `ImageObject` schema with broken `contentUrl` invalidates the entire article schema"; "The comparative table makes specific claims about competitor products; verify product details from current packaging / brand site to avoid misrepresentation"; "If retrieval bots are blocked at robots.txt, the moat asset cannot be cited — verify Technical SEO leaf 10 audit before publishing"; "FDA OTC drug claim language must be reviewed — Sun Drops cannot be marketed with medical / dermatological claims per SEO foundation goals and constraints." ## Open questions