[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M718MMZX71GMEHKT8AXAZS6P":3,"skill-same:01M718MMZX71GMEHKT8AXAZS6P":86},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":28,"publishedAt":49,"linkOnly":50,"markdownContent":51,"markdownBody":52,"frontMatter":53,"refName":54,"commitSha":55,"skillDirectory":56,"forksCount":57,"repositoryHtmlUrl":58,"repositoryHomepageUrl":6,"ownerHtmlUrl":59,"ownerBlogUrl":6,"ownerDescription":60,"updatedAt":61,"install":62,"resources":75},"01M718MMZX71GMEHKT8AXAZS6P","bofu-seo-aeo-strategy",null,"Plan, write, and maintain content that wins purchase-near buyers in both organic search and AI answer engines. BOFU content is structurally different from TOFU — the reader has accepted they need to decide; the page's job is to resolve the decision. Get it right and a single deep page can outperform dozens of informational posts on revenue contribution.","narayan-metaflow\u002Fmetaflow-marketing-skills","narayan-metaflow","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F181216517?v=4","01M718MB0Y27AYKTC9ZKT16V86","narayan","01M718MB209XRYRTS3ZZFK0JXS",19,"MIT","https:\u002F\u002Fgithub.com\u002Fnarayan-metaflow\u002Fmetaflow-marketing-skills\u002Fblob\u002Fmain\u002Fskills\u002Fbofu-seo-aeo-strategy\u002FSKILL.md",[18,23],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000019","营销推广","SCENARIO","使用场景",{"categoryNo":24,"categoryName":25,"taxonomyType":26,"taxonomyTypeDesc":27},"01JBSEEDCAT000000000000035","市场营销","PROFESSION","适用职业",[29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48],"seo","aeo","strategy","bofu seo aeo strategy","tofu","single","answer","page","write","revenue","plan","maintain","decision","purchase","search","decide","get","posts","deep","near","2026-11-10 10:12:30",false,"---\nname: bofu-seo-aeo-strategy\ndescription: \"Plan, write, and maintain content that wins purchase-near buyers in both organic search and AI answer engines. BOFU content is structurally different from TOFU — the reader has accepted they need to decide; the page's job is to resolve the decision. Get it right and a single deep page can outperform dozens of informational posts on revenue contribution.\"\n---\n\n## Why this matters\n\nBOFU queries — \"X vs Y\", \"alternatives to X\", \"best X\", \"best X for Y\" — are where late-stage evaluation happens. Buyers typing these have done their problem-awareness and solution-awareness work; they are choosing. These pages are also the most structurally standardized format in SEO, which means both Google's quality systems and AI answer engines have clear expectations of what a trustworthy, useful comparison or listicle looks like. Get the format right and credibility is compounding; get it wrong and the format itself flags the page as low-effort.\n\nTwo realities shape every BOFU decision:\n\n1. **Buyers filter BOFU content skeptically.** A third-party review platform or a peer Reddit thread carries more weight than a vendor's own comparison page. Your page must earn credibility inside a frame where it's presumed biased.\n2. **AI answer engines have reshuffled the citation pool.** Listicles, software-review aggregator pages, Reddit threads, YouTube review transcripts, and large publishers are disproportionately cited for commercial queries. Vendor-owned BOFU content alone does not earn AI citation share — third-party presence is a prerequisite, not a substitute.\n\nThis skill teaches the craft for producing BOFU content that (a) picks the right format for the intent, (b) earns credibility despite the conflict-of-interest frame, (c) is structured for both human scanning and LLM extraction, and (d) integrates with the third-party surfaces where buyers actually evaluate.\n\n## Process\n\n1. **Classify the query intent.** Pin the page to one of the four BOFU formats — comparison, alternative, listicle, or segment page. Format drives structure. Getting this wrong creates mid-funnel content wearing BOFU clothes.\n2. **Audit the SERP and citation pool.** Who ranks for the target query today? What do AI engines cite? Identify the content archetype (aggregator, publisher, vendor) dominating each surface.\n3. **Decide your angle and stance.** Will the page include your own product? If so, at what position and with what framing? Can you honestly win on the compared dimensions? What segments will you call out?\n4. **Gather first-hand material.** Use every tool you plan to compare. Capture screenshots, run representative workflows, note specific observations. BOFU credibility depends on demonstrable hands-on experience.\n5. **Structure the page to the canonical pattern.** Apply the format-specific H2 sequence (Section 3.2 below, and `references\u002Fformat-patterns.md` for depth).\n6. **Apply credibility moves.** Concession sections, worked pricing, dated screenshots, segmented verdicts, third-party ratings, explicit methodology. These are not optional; they are the difference between a trusted comparison and affiliate wallpaper.\n7. **Optimize for AI extraction.** Direct-answer first paragraph, question-form H2s, clean tables, fan-out sub-query coverage (FAQ section), entity density, summary box.\n8. **Invest in third-party presence in parallel.** software-review profiles with active reviews, legitimate Reddit participation, YouTube review presence, and brand-mention building. These are prerequisites, not afterthoughts.\n9. **Implement technically.** URL canonicalization, schema, internal linking from informational TOFU to the BOFU money page.\n10. **Set the maintenance cadence.** BOFU content decays faster than almost any other content type. Quarterly review is the floor.\n\n## Frameworks\n\n### 1. The four BOFU formats — distinguish and select\n\nMost BOFU mistakes happen before the first word is written, when the author treats all four formats as interchangeable. They are not.\n\n| Format | Typical query | Intent signal | Canonical structure | Default length |\n|---|---|---|---|---|\n| **Comparison page** | \"[Tool A] vs [Tool B]\", \"[Incumbent] vs [Challenger]\" | Active evaluation between 2–3 named tools | Feature Deep Dive (Pattern A) | 2,500–4,500 words |\n| **Alternative page** | \"Alternatives to [incumbent]\", \"[category] alternatives\" | Dissatisfaction with incumbent; open shortlist | Listicle with positioning (Pattern C) | 2,000–3,500 words |\n| **\"Best X\" listicle** | \"Best project management software\", \"Best CRM\" | Fresh evaluation across full category | Ranked listicle with methodology | 2,500–5,500 words |\n| **\"Best X for Y\" segment page** | \"Best CRM for small business\", \"Best PM tool for agencies\" | Segmented evaluation within a category | Segmented listicle with narrower scope | 2,000–3,500 words |\n\n**Rule:** one page = one format. Do not mix a comparison into a listicle or a listicle into a segment page. The H2 sequence and the reader's expectation are different for each.\n\n**Selecting between them:**\n\n- If the query names exactly two or three tools → **comparison**\n- If the query names one incumbent and asks for substitutes → **alternative**\n- If the query asks for the whole category → **listicle**\n- If the query narrows by audience, size, use case, or industry → **segment page**\n- If you are a vendor and the query is \"best [your category]\" → expect a head-term SERP dominated by review sites and large publishers; prefer **segment pages** where you can win on specificity\n\n### 2. Canonical structural patterns\n\nThese are the H2 sequences that recur across strong pages. Use them as scaffolding, not a straitjacket. See `references\u002Fformat-patterns.md` for deep patterns including per-tool sub-structures, table specifications, and worked H2 examples for each format.\n\n**Pattern A — Comparison page (X vs Y)**\n\n```\nH1: [Product A] vs [Product B]: [year] Comparison\nH2: [A] vs [B] at a glance  ← comparison table + TL;DR here\nH2: What is [A]?\nH2: What is [B]?\nH2: Feature comparison\n  H3: [Feature 1]\n  H3: [Feature 2]\n  H3: [Feature 3]\nH2: Pricing comparison  ← worked examples at specific volumes\nH2: User reviews and third-party ratings\nH2: Where [A] is the better choice\nH2: Where [B] is the better choice  ← the concession section\nH2: Which should you choose?  ← segmented recommendation\nH2: FAQ\n```\n\n**Pattern C — Alternative page and \"Best X\" listicle**\n\n```\nH1: [N] Best [Competitor] Alternatives \u002F Best [Category] in [year]\nH2: Why look for a [Competitor] alternative? \u002F Why this matters\nH2: What to look for  ← evaluation criteria\nH2: How we evaluated  ← methodology\nH2: Quick comparison  ← at-a-glance table\nH2: 1. [Tool] — Best for [use case]\n  H3: Key features\n  H3: Pricing\n  H3: Pros \u002F Cons\n  H3: Best for\nH2: 2. [Tool] — …\n  (7–12 tools total)\nH2: How to choose the right [tool]\nH2: FAQ\n```\n\n**Pattern D — Segment page (\"Best X for Y\")**\n\nUses Pattern C scaffolding with two adjustments: (a) evaluation criteria are segment-specific, not generic, and (b) the tool list is filtered to those that genuinely fit the segment. A \"best CRM for solo consultants\" page should not include enterprise-only tools, even if the parent \"best CRM\" page does.\n\n### 3. Credibility moves — the core craft\n\nA vendor or affiliated author writing BOFU content is read inside a presumed-bias frame. Credibility is earned through specific moves. These are non-negotiable on any page meant to rank or be cited by AI.\n\n**Objectivity moves:**\n- **Concession section.** At least one H2 acknowledging where a competitor is the better choice, with specifics (use cases, features, pricing tiers). Strong vendor-authored comparisons do this visibly—mirror that pattern.\n- **Segmented verdicts, not blanket winners.** \"A is better for enterprise; B is better for SMB\" instead of \"A is the winner.\" Recommendations split by use case.\n- **Third-party ratings cited** (software review aggregators and analyst summaries) rather than self-assessment.\n- **Explicit methodology** — how many tools were tested, over what period, against what criteria, by whom, with what disclosed connections.\n\n**Evidence moves:**\n- **Screenshots of the actual product in use** — dated, captioned, showing the specific workflow discussed. Not stock marketing images. Not the competitor's marketing site.\n- **Specific observations only possible from usage.** \"The CSV export truncates fields above 255 characters\" beats \"easy to export.\" The specificity is the proof.\n- **Worked pricing examples at specific volumes.** \"For a 25-person team on the Business plan, monthly cost: A = $X, B = $Y\" beats \"A is more expensive.\" Link to the competitor's official pricing page; timestamp your pricing data.\n- **Dated content.** Year in title, last-updated timestamp visible on-page.\n\n**Format moves:**\n- **Tables with descriptive cells, not binary checkmarks.** \"Unlimited users\" beats \"Yes\". \"Starts at $15\u002Fuser\u002Fmonth, annual only\" beats \"$$\".\n- **Named author with demonstrable expertise.** A LinkedIn-verifiable role, related writing history, or domain-credible background. Not \"Editorial Team\".\n- **Per-tool structure consistent across entries.** If Tool 1 gets Pricing \u002F Pros \u002F Cons \u002F Best for, every other tool gets the same, in the same order.\n\n**The self-inclusion question.** If your product is in the category, three stances exist:\n- **Include honestly, ranked on merit.** Acceptable if genuine first-hand testing of competitors, visible methodology, and no #1 position for yourself unless earned on the evaluation criteria.\n- **Exclude and link.** Keep the main listicle competitor-only; maintain a separate \"How we compare\" page for readers who want your view.\n- **Split.** Main content excludes; a clearly-disclosed vendor perspective box sits at the end.\n\nDefault stance when uncertain: **exclude and link, or split.** Ranking yourself #1 in your own list is a pattern Google systems and sophisticated buyers both detect.\n\n### 4. The \"ski ramp\" applied to BOFU\n\nThe ski ramp structure — front-loaded answer, question-form H2s, high entity density, definitive phrasing — applies to BOFU but with BOFU-specific adaptations. See the `seo-content-strategy` skill for the general framework; the adaptations for BOFU are:\n\n- **Direct-answer first paragraph is a segmented recommendation, not a single pick.** \"For most small teams, A is the strongest choice. If you need deep integrations, B. If budget is the main constraint, C.\" This mirrors how AI engines synthesize commercial answers.\n- **The TL;DR comparison table goes immediately under the intro.** It serves both the skimmer and the LLM extractor, which frequently pulls from early tables.\n- **H2s phrase the buyer's decision question.** \"Which is better for small teams?\" beats \"Team size considerations.\" \"How do A and B compare on pricing?\" beats \"Pricing breakdown.\"\n- **Entity density is naturally high in BOFU** — product names, feature names, pricing tiers, integrations, third-party review scores. Do not stuff; let the format carry the density.\n- **Summary box at the end** with the segmented verdict restated. AI engines sometimes pull the closing summary as the extractable answer.\n\n### 5. AI visibility beyond the page — third-party presence\n\nVendor-owned BOFU content alone does not earn AI citation share for commercial queries. Citation pools across ChatGPT, Perplexity, AI Overviews, AI Mode, Copilot, and Gemini converge on a similar set: large publisher listicles, review-site category pages, Reddit threads, YouTube review transcripts, and established review aggregators. See `references\u002Faeo-for-bofu.md` for platform-specific behavior and citation pool depth.\n\n**Non-content actions that now matter as much as the page itself:**\n\n- **Software review marketplace profiles with active recent reviews.** Profile completeness, recent review velocity, response to reviews. These are inputs to the AI citation pool.\n- **Legitimate Reddit presence.** Participate in two or three relevant subreddits with real substance. Answer questions without pitching. Do not seed fake discussions — both platforms and buyers detect it and the reputational cost is asymmetric.\n- **YouTube review presence.** Either your own channel with real walkthroughs, or earned coverage from reviewers. Transcripts enter the citation pool for ChatGPT and Gemini in particular.\n- **Brand mentions at volume.** Unlinked brand mentions in industry coverage, podcasts, and newsletters build the co-occurrence patterns LLMs retrieve on. PR and earned media have become AEO channels, not just branding channels.\n- **Podcast presence.** Guest appearances on podcasts with transcripts crawled and indexed.\n\nThe shape of the work changes: the comparison page is one surface, but the brand must be present across the citation pool or the page alone will underperform.\n\n### 6. Programmatic \"Best X for Y\" at scale\n\nSegment pages (\"best X for Y\") are the most scalable BOFU format — less competition than head terms, clearer intent, natural fit for a vendor's editorial voice. They are also the format most at risk of slipping into doorway \u002F thin content if scaled carelessly.\n\n**When a segment page deserves to exist:**\n- The segment (Y) is a real buying segment — buyers self-identify with that label, not a marketing construct\n- The tool recommendations genuinely differ from the parent category list (different tools, or the same tools in a different order, for segment-specific reasons)\n- You can write segment-specific evaluation criteria, not just filter the parent list\n- You have segment-specific research, case study, or audience insight to anchor the page\n\n**When a segment page is thin content:**\n- Headline swap with no substantive change from the parent page\n- Segment modifier that doesn't change the recommendations\n- Generated at volume without per-page insight\n- Same evaluation criteria as the head term, just with the Y inserted\n\n**Scaling pattern for responsible \"best X for Y\":**\n1. Identify real, high-intent Y modifiers from search data, sales conversations, and buyer interviews\n2. Cluster them into segments that actually change the answer\n3. Produce one page per segment, with unique: evaluation criteria, tool list ordering, use-case examples, and segment-specific insight\n4. Write (or review) every page by a human with visible authorship\n5. Cap the initial set at 8–20 segments; measure performance before expanding\n\nIf the plan is 200+ near-duplicate segment pages at launch, the work belongs in a true programmatic system with database-driven unique content per page — not a skill-driven editorial production line. Either invest in the per-page uniqueness or don't produce them.\n\n### 7. Technical implementation\n\n**URL structure:**\n- `\u002Fcompare\u002F[a]-vs-[b]\u002F` — neutral hub pattern, works if you publish many\n- `\u002F[your-product]-vs-[competitor]\u002F` — SEO-preferred when you authored it; signals authorship\n- `\u002Falternatives\u002F[competitor]\u002F` — distinct page type from vs-pages; can coexist\n- `\u002Fbest-[category]\u002F` and `\u002Fbest-[category]-for-[segment]\u002F` — clean, readable\n\n**Canonicalization:** publish one direction of X vs Y. If you are Tool A, write \"A vs B\" and canonicalize or 301 any Y vs X slug variations. Never publish both as indexable pages — duplicate content risk, split authority, both underperform.\n\n**Schema:**\n- `Article` or `BlogPosting` as page-level type\n- `FAQPage` for the FAQ section when present\n- `ItemList` + `Product` for alternatives and listicle formats\n- Avoid self-serving `Review` \u002F `AggregateRating` on your own product — Google has issued manual actions for this; use only third-party verified ratings\n\n**Internal linking:**\n- Link from informational TOFU and MOFU content on the topic into the BOFU page (the Siege Media pattern)\n- Cross-link related comparisons and alternatives where the topic overlaps\n- Link from the BOFU page back up to the parent pillar or category page\n- Use descriptive anchor text matching the target page's primary intent (\"compare A and B\", not \"click here\")\n\n### 8. Maintenance cadence\n\nBOFU content decays faster than almost any other content type. Competitor pricing changes, features ship, rebrands happen, your own product evolves. Bake maintenance into the workflow from day one.\n\n**Quarterly review is the floor for priority BOFU pages:**\n- Competitor pricing verification (link to source; update worked examples)\n- Feature accuracy pass (anything new since last review?)\n- Screenshot refresh where UI has changed substantially\n- Third-party rating \u002F review volume update\n- Year in title refresh at least annually\n\n**Ad-hoc triggers** that require immediate refresh:\n- Competitor pricing change, rebrand, acquisition, or major launch\n- Your own pricing, positioning, or feature change\n- Core Google update (particularly quality- or reviews-focused ones)\n- AI answer engine citation pattern shift (new sources surfacing)\n\n### 9. KPIs and measurement\n\nBOFU is measured differently from TOFU. The page's job is pipeline contribution, not raw traffic.\n\n| KPI | What it measures | Caveat |\n|---|---|---|\n| **Share of AI citations** across ChatGPT, Perplexity, AI Overviews, AI Mode, Copilot, Gemini | Brand presence in the citation pool for target commercial queries | AI citations are volatile; measure quarterly patterns, not day-to-day |\n| **Branded search lift** after the page ships | Proxy for unlinked brand mentions and AI-induced awareness | Hard to attribute cleanly; look for directional change |\n| **Assisted conversions** | Whether the page appears in buyer journeys that convert | Requires multi-touch attribution; last-click severely undercounts BOFU value |\n| **Third-party review velocity** | Rate of new user reviews on third-party platforms | Downstream of the page itself; tracks broader BOFU investment |\n| **Organic ranking position** for primary + segment queries | Classical SEO signal; still matters where AI Overviews don't dominate | Ranking without AI citation presence is increasingly a partial win |\n| **Time-to-shortlist** from sales data | Whether BOFU content is shortening the buyer's shortlisting step | Requires closing the loop with sales |\n\n**On the conversion-rate claim.** BOFU pages are often reported as converting at dramatically higher rates than TOFU. Read this carefully: it describes conversion conditional on reaching the page, not the page's persuasive power. Buyers arriving at a comparison page are overwhelmingly late-stage, already in-market, and influenced by Reddit, review sites, and peer referrals. The page verifies preferences; it does not create them. Budget accordingly.\n\n## Ethics and professional posture\n\nWrite BOFU content the way you would want a competitor to write about you: truthful, specific, fair. Substantiate factual claims about competitors. Link to their pricing page rather than paraphrasing it. Acknowledge where they do things well. Keep the tone professional — never disparaging, never anticompetitive in spirit, never exaggerating a weakness into a caricature. This is both an ethical posture and a credibility engine: buyers and AI engines alike reward pages that read as fair-minded, and penalize pages that read as hit pieces. For regulated categories (finance, health, legal, anything with a meaningful compliance surface), loop in counsel before publication.\n\n## Output format\n\n```\n## BOFU Content Brief — [Page title]\n\n**Format:** [Comparison \u002F Alternative \u002F Listicle \u002F Segment page]\n**Primary query:** [Exact target]\n**Fan-out sub-queries:** [3–8 related queries the page should cover]\n**Target word count:** [Based on format and SERP]\n\n### Strategic context\n- **SERP archetypes dominating this query:** [Aggregators \u002F publishers \u002F vendors \u002F Reddit]\n- **AI citation pool:** [Who gets cited across ChatGPT, Perplexity, AIO, AI Mode]\n- **Our angle \u002F differentiator:** [The specific POV or evidence this page adds]\n- **Self-inclusion stance:** [Include honestly \u002F exclude and link \u002F split]\n\n### Structural outline\n[Format-specific H2 sequence, with one-line summary per H2]\n\n### Credibility moves checklist\n- [ ] Concession section with specifics\n- [ ] Segmented verdict, not blanket pick\n- [ ] Dated screenshots from first-hand use\n- [ ] Worked pricing examples at specific volumes, timestamped\n- [ ] Third-party ratings cited\n- [ ] Explicit methodology section\n- [ ] Named author with credentials\n- [ ] Descriptive-cell comparison table (not binary checkmarks)\n\n### AI extraction checklist\n- [ ] Direct-answer first paragraph (segmented recommendation)\n- [ ] TL;DR table immediately under intro\n- [ ] Question-form H2s\n- [ ] FAQ covering fan-out sub-queries\n- [ ] Summary box at the end\n\n### Third-party presence plan\n- **Review sites:** [Profile status, review velocity plan]\n- **Reddit:** [Target subs, engagement approach]\n- **YouTube:** [Own content \u002F earned coverage plan]\n- **PR \u002F mentions:** [Target publications or newsletters]\n\n### Technical spec\n- URL: [Final slug]\n- Canonical direction: [If comparison, which direction is published]\n- Schema types: [Article + FAQPage + ItemList\u002FProduct as applicable]\n- Internal link sources: [Pages pointing to this one]\n\n### Maintenance\n- Review cadence: [Quarterly default]\n- Named owner: [Who maintains]\n- Refresh triggers documented: [Y\u002FN]\n\n### KPIs\n| KPI | Baseline | Target | Review |\n|---|---|---|---|\n```\n\n## Example — generic head-to-head comparison\n\n**Page:** \"[Your product] vs [Competitor]: [year] comparison\"  \n**Format:** Comparison page (Pattern A)  \n**Primary query:** exact head-to-head string buyers type  \n**Target word count:** driven by SERP depth (often 2,500–4,500 words)\n\n**Strategic context:** For direct `[A] vs [B]` queries, vendor-authored comparisons and review-site compare pages often dominate citations. Your angle: first-hand use, explicit concessions (“where [B] wins”), worked pricing with dates, and methodology.\n\n**Structural outline (pattern):**\n- H1: head-to-head title  \n- H2: at-a-glance table (pricing model, deployment, core capabilities, integrations, review scores, best-for)  \n- H2: factual overview of each product (no snark)  \n- H2: dimension-by-dimension comparison (subheads per buyer concern)  \n- H2: pricing with worked examples at 2+ volumes  \n- H2: third-party ratings (linked)  \n- H2: where each option is the better fit (segmented, not a blanket winner)  \n- H2: FAQ aligned to fan-out sub-queries  \n\n**Technical:** single canonical direction for the pair; `Article` + `FAQPage` as appropriate; no self-serving `Review` \u002F `AggregateRating` on your own product.\n\n**Maintenance:** quarterly pricing\u002Ffeature pass; refresh when either product ships materially; update year in title annually.\n\n## Guidelines\n\n- **Pick the format first, then write.** Comparison, alternative, listicle, and segment page are four different contracts with the reader. Do not mix.\n- **Assume presumed bias and earn credibility move by move.** Concession section, worked pricing, dated screenshots, segmented verdicts, third-party ratings, explicit methodology. These are the craft.\n- **Never rank yourself #1 in your own \"Best X\" list.** Default to exclude-and-link or split. If you include honestly, do not take the top slot.\n- **First-hand use is non-negotiable.** If nobody on the team has used Tool N, Tool N does not belong in a vendor-authored comparison or listicle. Google's reviews guidance and buyer trust both require it.\n- **Write the concession section specifically.** \"Tool B has some nice features\" is not a concession. \"Tool B's report builder handles multi-dimensional pivots that Tool A does not support, which matters for teams doing finance-grade attribution\" is.\n- **Segment verdicts over winner verdicts.** Blanket \"Tool A wins\" ignores real segment differences and reads as marketing. \"A for small teams; B for enterprise; C for open-source preference\" reads as useful.\n- **One direction per X vs Y.** Canonicalize. Split authority is a self-inflicted ranking wound.\n- **Tables: descriptive cells, HTML markup, not screenshots.** \"Unlimited users, starts at $15\u002Fuser\u002Fmo annual\" beats \"Yes \u002F $$$\". LLMs extract from HTML tables; image tables are invisible to them.\n- **Front-load the segmented recommendation.** The first paragraph should already be answering the buyer's decision question.\n- **Use question-form H2s.** \"Which is better for small teams?\" beats \"Team size considerations.\" LLMs treat H2s as prompts.\n- **Treat third-party presence as a prerequisite.** Review sites, Reddit, YouTube, PR. Vendor content alone does not earn AI citation share for commercial queries.\n- **\"Best X for Y\" at scale only if each page carries unique value.** Headline-swap segment pages are the format Google has repeatedly targeted. Either invest in per-page insight or do not publish the page.\n- **Schema honestly.** `Article` + `FAQPage` + `ItemList\u002FProduct` where appropriate. No self-serving `Review` or `AggregateRating` on your own product.\n- **Quarterly review is the floor, not the ceiling.** Competitor pricing, feature accuracy, screenshot freshness, year-in-title. Name an owner.\n- **Measure pipeline, not just traffic.** BOFU performance is qualified pipeline contribution. Rankings and visits are leading indicators; the conversion rate often cited for BOFU describes already-in-market buyers, not the page's persuasive power.\n- **Be truthful, fair, and professional in tone.** Never anticompetitive in spirit. Substantiate factual claims. Link to sources. Acknowledge competitor strengths with specifics. Buyers and AI engines both reward fair-minded writing.\n- **Loop in counsel for regulated categories.** Finance, health, legal, or anything with a meaningful compliance surface — get a second pair of eyes before publication.\n\n## Reference files\n\nWhen implementing a specific BOFU page or auditing an AI citation strategy, read the relevant reference:\n\n- `references\u002Fformat-patterns.md` — Deep structural patterns for each BOFU format, with canonical H2 sequences, per-tool sub-structure templates, comparison-table specifications, and worked H2 examples. Read when drafting or reviewing the page structure.\n- `references\u002Faeo-for-bofu.md` — Platform-specific AI answer engine behavior for commercial queries, citation pool depth, fan-out sub-query patterns for BOFU intents, and optimization moves for each platform. Read when planning for AI citation visibility or auditing current citation share.\n\n## Cross-skill handoffs\n\n- **← seo-keyword-research:** Receive the commercial-intent keyword map with fan-out sub-queries, segment modifiers, and difficulty scores.\n- **← seo-content-strategy:** Receive the content brief with ski ramp defaults. BOFU pages are usually high-priority pages inside the broader cluster architecture.\n- **→ seo-onpage-optimization:** Hand off the structured brief for on-page implementation (schema, internal linking, on-page ski ramp polish).\n- **→ seo-technical-audit:** Hand off URL canonicalization and schema implementation review.\n- **→ seo-link-building:** Hand off third-party presence plan — earned listicle placements, PR for brand mentions, review-site promotion.\n- **→ aeo-ai-search-visibility:** Hand off the citation-share tracking plan across AI platforms.\n- **→ seo-reporting:** Hand off the KPI dashboard — share of AI citations, branded search lift, assisted conversions, review velocity, ranking positions.\n","## Why this matters\n\nBOFU queries — \"X vs Y\", \"alternatives to X\", \"best X\", \"best X for Y\" — are where late-stage evaluation happens. Buyers typing these have done their problem-awareness and solution-awareness work; they are choosing. These pages are also the most structurally standardized format in SEO, which means both Google's quality systems and AI answer engines have clear expectations of what a trustworthy, useful comparison or listicle looks like. Get the format right and credibility is compounding; get it wrong and the format itself flags the page as low-effort.\n\nTwo realities shape every BOFU decision:\n\n1. **Buyers filter BOFU content skeptically.** A third-party review platform or a peer Reddit thread carries more weight than a vendor's own comparison page. Your page must earn credibility inside a frame where it's presumed biased.\n2. **AI answer engines have reshuffled the citation pool.** Listicles, software-review aggregator pages, Reddit threads, YouTube review transcripts, and large publishers are disproportionately cited for commercial queries. Vendor-owned BOFU content alone does not earn AI citation share — third-party presence is a prerequisite, not a substitute.\n\nThis skill teaches the craft for producing BOFU content that (a) picks the right format for the intent, (b) earns credibility despite the conflict-of-interest frame, (c) is structured for both human scanning and LLM extraction, and (d) integrates with the third-party surfaces where buyers actually evaluate.\n\n## Process\n\n1. **Classify the query intent.** Pin the page to one of the four BOFU formats — comparison, alternative, listicle, or segment page. Format drives structure. Getting this wrong creates mid-funnel content wearing BOFU clothes.\n2. **Audit the SERP and citation pool.** Who ranks for the target query today? What do AI engines cite? Identify the content archetype (aggregator, publisher, vendor) dominating each surface.\n3. **Decide your angle and stance.** Will the page include your own product? If so, at what position and with what framing? Can you honestly win on the compared dimensions? What segments will you call out?\n4. **Gather first-hand material.** Use every tool you plan to compare. Capture screenshots, run representative workflows, note specific observations. BOFU credibility depends on demonstrable hands-on experience.\n5. **Structure the page to the canonical pattern.** Apply the format-specific H2 sequence (Section 3.2 below, and `references\u002Fformat-patterns.md` for depth).\n6. **Apply credibility moves.** Concession sections, worked pricing, dated screenshots, segmented verdicts, third-party ratings, explicit methodology. These are not optional; they are the difference between a trusted comparison and affiliate wallpaper.\n7. **Optimize for AI extraction.** Direct-answer first paragraph, question-form H2s, clean tables, fan-out sub-query coverage (FAQ section), entity density, summary box.\n8. **Invest in third-party presence in parallel.** software-review profiles with active reviews, legitimate Reddit participation, YouTube review presence, and brand-mention building. These are prerequisites, not afterthoughts.\n9. **Implement technically.** URL canonicalization, schema, internal linking from informational TOFU to the BOFU money page.\n10. **Set the maintenance cadence.** BOFU content decays faster than almost any other content type. Quarterly review is the floor.\n\n## Frameworks\n\n### 1. The four BOFU formats — distinguish and select\n\nMost BOFU mistakes happen before the first word is written, when the author treats all four formats as interchangeable. They are not.\n\n| Format | Typical query | Intent signal | Canonical structure | Default length |\n|---|---|---|---|---|\n| **Comparison page** | \"[Tool A] vs [Tool B]\", \"[Incumbent] vs [Challenger]\" | Active evaluation between 2–3 named tools | Feature Deep Dive (Pattern A) | 2,500–4,500 words |\n| **Alternative page** | \"Alternatives to [incumbent]\", \"[category] alternatives\" | Dissatisfaction with incumbent; open shortlist | Listicle with positioning (Pattern C) | 2,000–3,500 words |\n| **\"Best X\" listicle** | \"Best project management software\", \"Best CRM\" | Fresh evaluation across full category | Ranked listicle with methodology | 2,500–5,500 words |\n| **\"Best X for Y\" segment page** | \"Best CRM for small business\", \"Best PM tool for agencies\" | Segmented evaluation within a category | Segmented listicle with narrower scope | 2,000–3,500 words |\n\n**Rule:** one page = one format. Do not mix a comparison into a listicle or a listicle into a segment page. The H2 sequence and the reader's expectation are different for each.\n\n**Selecting between them:**\n\n- If the query names exactly two or three tools → **comparison**\n- If the query names one incumbent and asks for substitutes → **alternative**\n- If the query asks for the whole category → **listicle**\n- If the query narrows by audience, size, use case, or industry → **segment page**\n- If you are a vendor and the query is \"best [your category]\" → expect a head-term SERP dominated by review sites and large publishers; prefer **segment pages** where you can win on specificity\n\n### 2. Canonical structural patterns\n\nThese are the H2 sequences that recur across strong pages. Use them as scaffolding, not a straitjacket. See `references\u002Fformat-patterns.md` for deep patterns including per-tool sub-structures, table specifications, and worked H2 examples for each format.\n\n**Pattern A — Comparison page (X vs Y)**\n\n```\nH1: [Product A] vs [Product B]: [year] Comparison\nH2: [A] vs [B] at a glance  ← comparison table + TL;DR here\nH2: What is [A]?\nH2: What is [B]?\nH2: Feature comparison\n  H3: [Feature 1]\n  H3: [Feature 2]\n  H3: [Feature 3]\nH2: Pricing comparison  ← worked examples at specific volumes\nH2: User reviews and third-party ratings\nH2: Where [A] is the better choice\nH2: Where [B] is the better choice  ← the concession section\nH2: Which should you choose?  ← segmented recommendation\nH2: FAQ\n```\n\n**Pattern C — Alternative page and \"Best X\" listicle**\n\n```\nH1: [N] Best [Competitor] Alternatives \u002F Best [Category] in [year]\nH2: Why look for a [Competitor] alternative? \u002F Why this matters\nH2: What to look for  ← evaluation criteria\nH2: How we evaluated  ← methodology\nH2: Quick comparison  ← at-a-glance table\nH2: 1. [Tool] — Best for [use case]\n  H3: Key features\n  H3: Pricing\n  H3: Pros \u002F Cons\n  H3: Best for\nH2: 2. [Tool] — …\n  (7–12 tools total)\nH2: How to choose the right [tool]\nH2: FAQ\n```\n\n**Pattern D — Segment page (\"Best X for Y\")**\n\nUses Pattern C scaffolding with two adjustments: (a) evaluation criteria are segment-specific, not generic, and (b) the tool list is filtered to those that genuinely fit the segment. A \"best CRM for solo consultants\" page should not include enterprise-only tools, even if the parent \"best CRM\" page does.\n\n### 3. Credibility moves — the core craft\n\nA vendor or affiliated author writing BOFU content is read inside a presumed-bias frame. Credibility is earned through specific moves. These are non-negotiable on any page meant to rank or be cited by AI.\n\n**Objectivity moves:**\n- **Concession section.** At least one H2 acknowledging where a competitor is the better choice, with specifics (use cases, features, pricing tiers). Strong vendor-authored comparisons do this visibly—mirror that pattern.\n- **Segmented verdicts, not blanket winners.** \"A is better for enterprise; B is better for SMB\" instead of \"A is the winner.\" Recommendations split by use case.\n- **Third-party ratings cited** (software review aggregators and analyst summaries) rather than self-assessment.\n- **Explicit methodology** — how many tools were tested, over what period, against what criteria, by whom, with what disclosed connections.\n\n**Evidence moves:**\n- **Screenshots of the actual product in use** — dated, captioned, showing the specific workflow discussed. Not stock marketing images. Not the competitor's marketing site.\n- **Specific observations only possible from usage.** \"The CSV export truncates fields above 255 characters\" beats \"easy to export.\" The specificity is the proof.\n- **Worked pricing examples at specific volumes.** \"For a 25-person team on the Business plan, monthly cost: A = $X, B = $Y\" beats \"A is more expensive.\" Link to the competitor's official pricing page; timestamp your pricing data.\n- **Dated content.** Year in title, last-updated timestamp visible on-page.\n\n**Format moves:**\n- **Tables with descriptive cells, not binary checkmarks.** \"Unlimited users\" beats \"Yes\". \"Starts at $15\u002Fuser\u002Fmonth, annual only\" beats \"$$\".\n- **Named author with demonstrable expertise.** A LinkedIn-verifiable role, related writing history, or domain-credible background. Not \"Editorial Team\".\n- **Per-tool structure consistent across entries.** If Tool 1 gets Pricing \u002F Pros \u002F Cons \u002F Best for, every other tool gets the same, in the same order.\n\n**The self-inclusion question.** If your product is in the category, three stances exist:\n- **Include honestly, ranked on merit.** Acceptable if genuine first-hand testing of competitors, visible methodology, and no #1 position for yourself unless earned on the evaluation criteria.\n- **Exclude and link.** Keep the main listicle competitor-only; maintain a separate \"How we compare\" page for readers who want your view.\n- **Split.** Main content excludes; a clearly-disclosed vendor perspective box sits at the end.\n\nDefault stance when uncertain: **exclude and link, or split.** Ranking yourself #1 in your own list is a pattern Google systems and sophisticated buyers both detect.\n\n### 4. The \"ski ramp\" applied to BOFU\n\nThe ski ramp structure — front-loaded answer, question-form H2s, high entity density, definitive phrasing — applies to BOFU but with BOFU-specific adaptations. See the `seo-content-strategy` skill for the general framework; the adaptations for BOFU are:\n\n- **Direct-answer first paragraph is a segmented recommendation, not a single pick.** \"For most small teams, A is the strongest choice. If you need deep integrations, B. If budget is the main constraint, C.\" This mirrors how AI engines synthesize commercial answers.\n- **The TL;DR comparison table goes immediately under the intro.** It serves both the skimmer and the LLM extractor, which frequently pulls from early tables.\n- **H2s phrase the buyer's decision question.** \"Which is better for small teams?\" beats \"Team size considerations.\" \"How do A and B compare on pricing?\" beats \"Pricing breakdown.\"\n- **Entity density is naturally high in BOFU** — product names, feature names, pricing tiers, integrations, third-party review scores. Do not stuff; let the format carry the density.\n- **Summary box at the end** with the segmented verdict restated. AI engines sometimes pull the closing summary as the extractable answer.\n\n### 5. AI visibility beyond the page — third-party presence\n\nVendor-owned BOFU content alone does not earn AI citation share for commercial queries. Citation pools across ChatGPT, Perplexity, AI Overviews, AI Mode, Copilot, and Gemini converge on a similar set: large publisher listicles, review-site category pages, Reddit threads, YouTube review transcripts, and established review aggregators. See `references\u002Faeo-for-bofu.md` for platform-specific behavior and citation pool depth.\n\n**Non-content actions that now matter as much as the page itself:**\n\n- **Software review marketplace profiles with active recent reviews.** Profile completeness, recent review velocity, response to reviews. These are inputs to the AI citation pool.\n- **Legitimate Reddit presence.** Participate in two or three relevant subreddits with real substance. Answer questions without pitching. Do not seed fake discussions — both platforms and buyers detect it and the reputational cost is asymmetric.\n- **YouTube review presence.** Either your own channel with real walkthroughs, or earned coverage from reviewers. Transcripts enter the citation pool for ChatGPT and Gemini in particular.\n- **Brand mentions at volume.** Unlinked brand mentions in industry coverage, podcasts, and newsletters build the co-occurrence patterns LLMs retrieve on. PR and earned media have become AEO channels, not just branding channels.\n- **Podcast presence.** Guest appearances on podcasts with transcripts crawled and indexed.\n\nThe shape of the work changes: the comparison page is one surface, but the brand must be present across the citation pool or the page alone will underperform.\n\n### 6. Programmatic \"Best X for Y\" at scale\n\nSegment pages (\"best X for Y\") are the most scalable BOFU format — less competition than head terms, clearer intent, natural fit for a vendor's editorial voice. They are also the format most at risk of slipping into doorway \u002F thin content if scaled carelessly.\n\n**When a segment page deserves to exist:**\n- The segment (Y) is a real buying segment — buyers self-identify with that label, not a marketing construct\n- The tool recommendations genuinely differ from the parent category list (different tools, or the same tools in a different order, for segment-specific reasons)\n- You can write segment-specific evaluation criteria, not just filter the parent list\n- You have segment-specific research, case study, or audience insight to anchor the page\n\n**When a segment page is thin content:**\n- Headline swap with no substantive change from the parent page\n- Segment modifier that doesn't change the recommendations\n- Generated at volume without per-page insight\n- Same evaluation criteria as the head term, just with the Y inserted\n\n**Scaling pattern for responsible \"best X for Y\":**\n1. Identify real, high-intent Y modifiers from search data, sales conversations, and buyer interviews\n2. Cluster them into segments that actually change the answer\n3. Produce one page per segment, with unique: evaluation criteria, tool list ordering, use-case examples, and segment-specific insight\n4. Write (or review) every page by a human with visible authorship\n5. Cap the initial set at 8–20 segments; measure performance before expanding\n\nIf the plan is 200+ near-duplicate segment pages at launch, the work belongs in a true programmatic system with database-driven unique content per page — not a skill-driven editorial production line. Either invest in the per-page uniqueness or don't produce them.\n\n### 7. Technical implementation\n\n**URL structure:**\n- `\u002Fcompare\u002F[a]-vs-[b]\u002F` — neutral hub pattern, works if you publish many\n- `\u002F[your-product]-vs-[competitor]\u002F` — SEO-preferred when you authored it; signals authorship\n- `\u002Falternatives\u002F[competitor]\u002F` — distinct page type from vs-pages; can coexist\n- `\u002Fbest-[category]\u002F` and `\u002Fbest-[category]-for-[segment]\u002F` — clean, readable\n\n**Canonicalization:** publish one direction of X vs Y. If you are Tool A, write \"A vs B\" and canonicalize or 301 any Y vs X slug variations. Never publish both as indexable pages — duplicate content risk, split authority, both underperform.\n\n**Schema:**\n- `Article` or `BlogPosting` as page-level type\n- `FAQPage` for the FAQ section when present\n- `ItemList` + `Product` for alternatives and listicle formats\n- Avoid self-serving `Review` \u002F `AggregateRating` on your own product — Google has issued manual actions for this; use only third-party verified ratings\n\n**Internal linking:**\n- Link from informational TOFU and MOFU content on the topic into the BOFU page (the Siege Media pattern)\n- Cross-link related comparisons and alternatives where the topic overlaps\n- Link from the BOFU page back up to the parent pillar or category page\n- Use descriptive anchor text matching the target page's primary intent (\"compare A and B\", not \"click here\")\n\n### 8. Maintenance cadence\n\nBOFU content decays faster than almost any other content type. Competitor pricing changes, features ship, rebrands happen, your own product evolves. Bake maintenance into the workflow from day one.\n\n**Quarterly review is the floor for priority BOFU pages:**\n- Competitor pricing verification (link to source; update worked examples)\n- Feature accuracy pass (anything new since last review?)\n- Screenshot refresh where UI has changed substantially\n- Third-party rating \u002F review volume update\n- Year in title refresh at least annually\n\n**Ad-hoc triggers** that require immediate refresh:\n- Competitor pricing change, rebrand, acquisition, or major launch\n- Your own pricing, positioning, or feature change\n- Core Google update (particularly quality- or reviews-focused ones)\n- AI answer engine citation pattern shift (new sources surfacing)\n\n### 9. KPIs and measurement\n\nBOFU is measured differently from TOFU. The page's job is pipeline contribution, not raw traffic.\n\n| KPI | What it measures | Caveat |\n|---|---|---|\n| **Share of AI citations** across ChatGPT, Perplexity, AI Overviews, AI Mode, Copilot, Gemini | Brand presence in the citation pool for target commercial queries | AI citations are volatile; measure quarterly patterns, not day-to-day |\n| **Branded search lift** after the page ships | Proxy for unlinked brand mentions and AI-induced awareness | Hard to attribute cleanly; look for directional change |\n| **Assisted conversions** | Whether the page appears in buyer journeys that convert | Requires multi-touch attribution; last-click severely undercounts BOFU value |\n| **Third-party review velocity** | Rate of new user reviews on third-party platforms | Downstream of the page itself; tracks broader BOFU investment |\n| **Organic ranking position** for primary + segment queries | Classical SEO signal; still matters where AI Overviews don't dominate | Ranking without AI citation presence is increasingly a partial win |\n| **Time-to-shortlist** from sales data | Whether BOFU content is shortening the buyer's shortlisting step | Requires closing the loop with sales |\n\n**On the conversion-rate claim.** BOFU pages are often reported as converting at dramatically higher rates than TOFU. Read this carefully: it describes conversion conditional on reaching the page, not the page's persuasive power. Buyers arriving at a comparison page are overwhelmingly late-stage, already in-market, and influenced by Reddit, review sites, and peer referrals. The page verifies preferences; it does not create them. Budget accordingly.\n\n## Ethics and professional posture\n\nWrite BOFU content the way you would want a competitor to write about you: truthful, specific, fair. Substantiate factual claims about competitors. Link to their pricing page rather than paraphrasing it. Acknowledge where they do things well. Keep the tone professional — never disparaging, never anticompetitive in spirit, never exaggerating a weakness into a caricature. This is both an ethical posture and a credibility engine: buyers and AI engines alike reward pages that read as fair-minded, and penalize pages that read as hit pieces. For regulated categories (finance, health, legal, anything with a meaningful compliance surface), loop in counsel before publication.\n\n## Output format\n\n```\n## BOFU Content Brief — [Page title]\n\n**Format:** [Comparison \u002F Alternative \u002F Listicle \u002F Segment page]\n**Primary query:** [Exact target]\n**Fan-out sub-queries:** [3–8 related queries the page should cover]\n**Target word count:** [Based on format and SERP]\n\n### Strategic context\n- **SERP archetypes dominating this query:** [Aggregators \u002F publishers \u002F vendors \u002F Reddit]\n- **AI citation pool:** [Who gets cited across ChatGPT, Perplexity, AIO, AI Mode]\n- **Our angle \u002F differentiator:** [The specific POV or evidence this page adds]\n- **Self-inclusion stance:** [Include honestly \u002F exclude and link \u002F split]\n\n### Structural outline\n[Format-specific H2 sequence, with one-line summary per H2]\n\n### Credibility moves checklist\n- [ ] Concession section with specifics\n- [ ] Segmented verdict, not blanket pick\n- [ ] Dated screenshots from first-hand use\n- [ ] Worked pricing examples at specific volumes, timestamped\n- [ ] Third-party ratings cited\n- [ ] Explicit methodology section\n- [ ] Named author with credentials\n- [ ] Descriptive-cell comparison table (not binary checkmarks)\n\n### AI extraction checklist\n- [ ] Direct-answer first paragraph (segmented recommendation)\n- [ ] TL;DR table immediately under intro\n- [ ] Question-form H2s\n- [ ] FAQ covering fan-out sub-queries\n- [ ] Summary box at the end\n\n### Third-party presence plan\n- **Review sites:** [Profile status, review velocity plan]\n- **Reddit:** [Target subs, engagement approach]\n- **YouTube:** [Own content \u002F earned coverage plan]\n- **PR \u002F mentions:** [Target publications or newsletters]\n\n### Technical spec\n- URL: [Final slug]\n- Canonical direction: [If comparison, which direction is published]\n- Schema types: [Article + FAQPage + ItemList\u002FProduct as applicable]\n- Internal link sources: [Pages pointing to this one]\n\n### Maintenance\n- Review cadence: [Quarterly default]\n- Named owner: [Who maintains]\n- Refresh triggers documented: [Y\u002FN]\n\n### KPIs\n| KPI | Baseline | Target | Review |\n|---|---|---|---|\n```\n\n## Example — generic head-to-head comparison\n\n**Page:** \"[Your product] vs [Competitor]: [year] comparison\"  \n**Format:** Comparison page (Pattern A)  \n**Primary query:** exact head-to-head string buyers type  \n**Target word count:** driven by SERP depth (often 2,500–4,500 words)\n\n**Strategic context:** For direct `[A] vs [B]` queries, vendor-authored comparisons and review-site compare pages often dominate citations. Your angle: first-hand use, explicit concessions (“where [B] wins”), worked pricing with dates, and methodology.\n\n**Structural outline (pattern):**\n- H1: head-to-head title  \n- H2: at-a-glance table (pricing model, deployment, core capabilities, integrations, review scores, best-for)  \n- H2: factual overview of each product (no snark)  \n- H2: dimension-by-dimension comparison (subheads per buyer concern)  \n- H2: pricing with worked examples at 2+ volumes  \n- H2: third-party ratings (linked)  \n- H2: where each option is the better fit (segmented, not a blanket winner)  \n- H2: FAQ aligned to fan-out sub-queries  \n\n**Technical:** single canonical direction for the pair; `Article` + `FAQPage` as appropriate; no self-serving `Review` \u002F `AggregateRating` on your own product.\n\n**Maintenance:** quarterly pricing\u002Ffeature pass; refresh when either product ships materially; update year in title annually.\n\n## Guidelines\n\n- **Pick the format first, then write.** Comparison, alternative, listicle, and segment page are four different contracts with the reader. Do not mix.\n- **Assume presumed bias and earn credibility move by move.** Concession section, worked pricing, dated screenshots, segmented verdicts, third-party ratings, explicit methodology. These are the craft.\n- **Never rank yourself #1 in your own \"Best X\" list.** Default to exclude-and-link or split. If you include honestly, do not take the top slot.\n- **First-hand use is non-negotiable.** If nobody on the team has used Tool N, Tool N does not belong in a vendor-authored comparison or listicle. Google's reviews guidance and buyer trust both require it.\n- **Write the concession section specifically.** \"Tool B has some nice features\" is not a concession. \"Tool B's report builder handles multi-dimensional pivots that Tool A does not support, which matters for teams doing finance-grade attribution\" is.\n- **Segment verdicts over winner verdicts.** Blanket \"Tool A wins\" ignores real segment differences and reads as marketing. \"A for small teams; B for enterprise; C for open-source preference\" reads as useful.\n- **One direction per X vs Y.** Canonicalize. Split authority is a self-inflicted ranking wound.\n- **Tables: descriptive cells, HTML markup, not screenshots.** \"Unlimited users, starts at $15\u002Fuser\u002Fmo annual\" beats \"Yes \u002F $$$\". LLMs extract from HTML tables; image tables are invisible to them.\n- **Front-load the segmented recommendation.** The first paragraph should already be answering the buyer's decision question.\n- **Use question-form H2s.** \"Which is better for small teams?\" beats \"Team size considerations.\" LLMs treat H2s as prompts.\n- **Treat third-party presence as a prerequisite.** Review sites, Reddit, YouTube, PR. Vendor content alone does not earn AI citation share for commercial queries.\n- **\"Best X for Y\" at scale only if each page carries unique value.** Headline-swap segment pages are the format Google has repeatedly targeted. Either invest in per-page insight or do not publish the page.\n- **Schema honestly.** `Article` + `FAQPage` + `ItemList\u002FProduct` where appropriate. No self-serving `Review` or `AggregateRating` on your own product.\n- **Quarterly review is the floor, not the ceiling.** Competitor pricing, feature accuracy, screenshot freshness, year-in-title. Name an owner.\n- **Measure pipeline, not just traffic.** BOFU performance is qualified pipeline contribution. Rankings and visits are leading indicators; the conversion rate often cited for BOFU describes already-in-market buyers, not the page's persuasive power.\n- **Be truthful, fair, and professional in tone.** Never anticompetitive in spirit. Substantiate factual claims. Link to sources. Acknowledge competitor strengths with specifics. Buyers and AI engines both reward fair-minded writing.\n- **Loop in counsel for regulated categories.** Finance, health, legal, or anything with a meaningful compliance surface — get a second pair of eyes before publication.\n\n## Reference files\n\nWhen implementing a specific BOFU page or auditing an AI citation strategy, read the relevant reference:\n\n- `references\u002Fformat-patterns.md` — Deep structural patterns for each BOFU format, with canonical H2 sequences, per-tool sub-structure templates, comparison-table specifications, and worked H2 examples. Read when drafting or reviewing the page structure.\n- `references\u002Faeo-for-bofu.md` — Platform-specific AI answer engine behavior for commercial queries, citation pool depth, fan-out sub-query patterns for BOFU intents, and optimization moves for each platform. Read when planning for AI citation visibility or auditing current citation share.\n\n## Cross-skill handoffs\n\n- **← seo-keyword-research:** Receive the commercial-intent keyword map with fan-out sub-queries, segment modifiers, and difficulty scores.\n- **← seo-content-strategy:** Receive the content brief with ski ramp defaults. BOFU pages are usually high-priority pages inside the broader cluster architecture.\n- **→ seo-onpage-optimization:** Hand off the structured brief for on-page implementation (schema, internal linking, on-page ski ramp polish).\n- **→ seo-technical-audit:** Hand off URL canonicalization and schema implementation review.\n- **→ seo-link-building:** Hand off third-party presence plan — earned listicle placements, PR for brand mentions, review-site promotion.\n- **→ aeo-ai-search-visibility:** Hand off the citation-share tracking plan across AI platforms.\n- **→ seo-reporting:** Hand off the KPI dashboard — share of AI citations, branded search lift, assisted conversions, review velocity, ranking positions.\n",{"name":5,"description":7},"main","c137446804eb85b2cc0ba009c62068253a7c47be","skills\u002Fbofu-seo-aeo-strategy",2,"https:\u002F\u002Fgithub.com\u002Fnarayan-metaflow\u002Fmetaflow-marketing-skills","https:\u002F\u002Fgithub.com\u002Fnarayan-metaflow","engineer by education, marketer by profession, building rn","2026-09-16 13:41:44",{"source":8,"skillName":5,"commands":63,"prompt":73,"zipAvailable":74},[64,67,70],{"tool":65,"command":66},"npx","npx skills add narayan-metaflow\u002Fmetaflow-marketing-skills --skill bofu-seo-aeo-strategy",{"tool":68,"command":69},"bunx","bunx skills add narayan-metaflow\u002Fmetaflow-marketing-skills --skill bofu-seo-aeo-strategy",{"tool":71,"command":72},"pnpm","pnpm dlx skills add narayan-metaflow\u002Fmetaflow-marketing-skills --skill bofu-seo-aeo-strategy","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fnarayan-metaflow\u002Fmetaflow-marketing-skills\u002Fblob\u002Fmain\u002Fskills\u002Fbofu-seo-aeo-strategy\u002FSKILL.md",true,[76,82],{"path":77,"resourceType":78,"resourceTypeDesc":79,"sizeBytes":80,"sourceHtmlUrl":81},"references\u002Faeo-for-bofu.md","REFERENCE","参考文档",23915,"https:\u002F\u002Fgithub.com\u002Fnarayan-metaflow\u002Fmetaflow-marketing-skills\u002Fblob\u002Fmain\u002Fskills\u002Fbofu-seo-aeo-strategy\u002Freferences\u002Faeo-for-bofu.md",{"path":83,"resourceType":78,"resourceTypeDesc":79,"sizeBytes":84,"sourceHtmlUrl":85},"references\u002Fformat-patterns.md",16798,"https:\u002F\u002Fgithub.com\u002Fnarayan-metaflow\u002Fmetaflow-marketing-skills\u002Fblob\u002Fmain\u002Fskills\u002Fbofu-seo-aeo-strategy\u002Freferences\u002Fformat-patterns.md",[87,112,141,161,189,216],{"skillNo":88,"name":89,"nameZh":6,"description":90,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":91,"categories":92,"keywords":95,"publishedAt":49},"01M718Q16XB7D7ZMGNNA1226G5","technical-seo-ai-crawler-audit","Diagnose and prioritize the technical issues blocking a site from being crawled, indexed, ranked, and cited. Modern technical SEO has two audiences: Google's crawler\u002Findexer and AI answer engines (GPTBot, PerplexityBot, ClaudeBot, etc.). Both need HTML-visible content and a clean crawl path.","https:\u002F\u002Fgithub.com\u002Fnarayan-metaflow\u002Fmetaflow-marketing-skills\u002Fblob\u002Fmain\u002Fskills\u002Ftechnical-seo-ai-crawler-audit\u002FSKILL.md",[93,94],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":24,"categoryName":25,"taxonomyType":26,"taxonomyTypeDesc":27},[29,96,97,98,99,100,101,102,35,103,104,105,106,107,108,109,110,111],"technical","technical seo","technical seo ai crawler audit","audit","crawler","crawl","diagnose","issues","clean","indexer","google","prioritize","site","content","html","path",{"skillNo":113,"name":114,"nameZh":6,"description":115,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":116,"categories":117,"keywords":123,"publishedAt":49},"01M718PY2PKA7BYS5KNJ2A9DRT","seo-reporting-measurement","Build reports that stakeholders trust, rooted in data sources that are actually reliable. Modern SEO measurement has to reckon with a very specific constraint: **the primary data source (GSC) is ~75% incomplete**, and the AI visibility layer is **probabilistic**, not deterministic. Building reports that ignore these realities produces false confidence.","https:\u002F\u002Fgithub.com\u002Fnarayan-metaflow\u002Fmetaflow-marketing-skills\u002Fblob\u002Fmain\u002Fskills\u002Fseo-reporting-measurement\u002FSKILL.md",[118,119,122],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":120,"categoryName":121,"taxonomyType":26,"taxonomyTypeDesc":27},"01JBSEEDCAT000000000000033","数据分析师",{"categoryNo":24,"categoryName":25,"taxonomyType":26,"taxonomyTypeDesc":27},[124,125,126,127,29,128,129,130,131,132,133,134,135,136,137,138,139,140],"seo reporting","seo reporting measurement","reporting","measurement","sources","gsc","false","reports","data sources","data","layer","source","deterministic","trust","with","build","constraint",{"skillNo":142,"name":143,"nameZh":6,"description":144,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":145,"categories":146,"keywords":149,"publishedAt":49},"01M718PVF0SP35J6S9EHRGQZQ1","seo-max-kt","Long-form SEO and AEO methodology bundle; read SEO_AND_AEO_KT.md for the full corpus.","https:\u002F\u002Fgithub.com\u002Fnarayan-metaflow\u002Fmetaflow-marketing-skills\u002Fblob\u002Fmain\u002Fskills\u002Fseo-max-kt\u002FSKILL.md",[147,148],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":24,"categoryName":25,"taxonomyType":26,"taxonomyTypeDesc":27},[150,29,151,152,30,153,154,155,156,157,158,159,160],"kt","seo max kt","max","long-form","methodology","form","for","bundle","full","read","long",{"skillNo":162,"name":163,"nameZh":6,"description":164,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":165,"categories":166,"keywords":172,"publishedAt":49},"01M718PRQNX2JF71GBFTH3MV19","seo-keyword-research-intent-mapping","Build a keyword universe that reflects how both humans *and* AI answer engines search. Traditional keyword research looks at standalone terms; modern research also maps the **fan-out sub-queries** AI tools generate when breaking down a user prompt.","https:\u002F\u002Fgithub.com\u002Fnarayan-metaflow\u002Fmetaflow-marketing-skills\u002Fblob\u002Fmain\u002Fskills\u002Fseo-keyword-research-intent-mapping\u002FSKILL.md",[167,168,171],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":169,"categoryName":170,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000021","资料检索与研究",{"categoryNo":24,"categoryName":25,"taxonomyType":26,"taxonomyTypeDesc":27},[29,173,174,175,176,177,178,179,180,181,35,43,182,183,184,185,186,139,187,188],"seo keyword research","seo keyword research intent mapping","research","intent","keyword research","mapping","ai tools","standalone","down","prompt","when","maps","queries","generate","how","at",{"skillNo":190,"name":191,"nameZh":6,"description":192,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":193,"categories":194,"keywords":203,"publishedAt":49},"01M718PNDT3A19SV7KVWC9083P","seo-aeo-content-strategy","Design content systems that earn both Google rankings *and* AI citations. The core insight: AI answer engines read content differently than humans — they extract the highest-information-gain sentences and treat H2 headings as prompts. The \"ski ramp\" structure optimizes for both.","https:\u002F\u002Fgithub.com\u002Fnarayan-metaflow\u002Fmetaflow-marketing-skills\u002Fblob\u002Fmain\u002Fskills\u002Fseo-aeo-content-strategy\u002FSKILL.md",[195,198,199,200],{"categoryNo":196,"categoryName":197,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000015","视觉设计",{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":24,"categoryName":25,"taxonomyType":26,"taxonomyTypeDesc":27},{"categoryNo":201,"categoryName":202,"taxonomyType":26,"taxonomyTypeDesc":27},"01JBSEEDCAT000000000000036","内容创作者",[29,30,31,204,205,109,206,207,208,156,35,209,210,211,212,159,213,214,106,215],"seo aeo content strategy","content strategy","Citations","as","core","design","information","earn","structure","systems","ramp","prompts",{"skillNo":217,"name":218,"nameZh":6,"description":219,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":220,"categories":221,"keywords":225,"publishedAt":49},"01M718PJM103BE7581G6NRCC7T","prompt-picker","Build a prompt library that reveals how a brand shows up across the full buyer journey in AI search systems. The library is designed for actual LLM visibility tracking, not theoretical SEO keyword research.","https:\u002F\u002Fgithub.com\u002Fnarayan-metaflow\u002Fmetaflow-marketing-skills\u002Fblob\u002Fmain\u002Fskills\u002Fprompt-picker\u002FSKILL.md",[222],{"categoryNo":223,"categoryName":224,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发",[226,182,227,228,29,156,173,43,229,230,231,232,175,139,187,213,177,233,234,158],"prompt picker","llm visibility","brand","up","journey","prompt library","across","library","tracking"]