Skip to content
AEO Toolkit AEO Toolkit

SEO + AEO plan

The site must dogfood the product. AEO Toolkit teaches sites how to rank in classic search and in AI answer engines — so aeotoolkit itself ships llms.txt, JSON-LD, server-rendered semantic HTML, fast Core Web Vitals, and answer-first content. This plan covers both SEO (Google/Bing blue links) and AEO/GEO (being cited inside ChatGPT, Claude, Perplexity, Gemini, and Google AI Overviews).

0. Targets & thesis

  • Primary keywords: “answer engine optimization”, “AEO tools”, “llms.txt generator”, “AI SEO”, “E-E-A-T checker”, “GEO optimization”, “AI visibility tracker”, “technical SEO audit”.
  • AEO thesis: answer engines cite content that is (a) crawlable & structured, (b) demonstrably trustworthy (E-E-A-T), (c) written as direct, extractable answers, and (d) entity-consistent. We optimize for citation, not just ranking.

1. Technical SEO foundation (ship now)

ItemImplementation
Server-rendered HTMLNext.js App Router, SSR/SSG. Tool UIs are client, but each page ships meaningful static HTML (hero, copy, FAQ) so crawlers + AI bots get content without JS.
Canonical URLsmetadataBase + per-page alternates.canonical.
Sitemapsrc/app/sitemap.ts → /sitemap.xml listing the landing + every tool page + key anchors.
robots.txtsrc/app/robots.ts — allow all crawlers, explicitly allow AI bots (GPTBot, OAI-SearchBot, ChatGPT-User, ClaudeBot, anthropic-ai, PerplexityBot, Google-Extended), link the sitemap.
llms.txt/llms.txt route — the de-facto AI crawl-hint file (we generate it with our own tool).
HTTPS / HSTSVercel default.
Core Web VitalsFirst paint is CSS-only (the hero shader and node field load after hydration and are gated on prefers-reduced-motion); three.js dynamic-imported only on /tools/graph; next/font (no layout shift); image-free hero; lazy-load below-the-fold. Target LCP < 2.0s, CLS < 0.05, INP < 200ms.
MobileResponsive at 360/768/1280; tap targets ≥ 40px; viewport meta via Next.
MetadataTitle template, unique title + description per page, OG + Twitter cards, opengraph-image generated with next/og.

2. On-page SEO (per page)

  • One <h1> per page, descriptive; logical <h2>/<h3> outline (audit checkers love this — we literally score it).
  • Semantic HTML: <header> <main> <section> <article> <nav> <footer>, lists for steps, <table> for comparisons, <figure> for the graph.
  • Internal linking: landing → each tool; each tool → 2–3 sibling tools + the glossary; footer hub; descriptive anchor text (“run a free technical SEO audit”, not “click here”).
  • Descriptions: 70–160 chars, benefit-led, include the primary keyword once, end with a verb.
  • Image alt + dimensions on any imagery/OG.

3. Structured data (JSON-LD) — the AEO backbone

Rendered via @/components/seo/JsonLd. Per-page schema:

PageSchema types
Layout (all)Organization (Advance Labs, logo, sameAs GitHub) + WebSite (+ SearchAction)
LandingSoftwareApplication (AEO Toolkit, applicationCategory BusinessApplication, free offers) + FAQPage
Each tool pageWebPage + BreadcrumbList + a tool-specific FAQPage (3–5 Q&As) + HowTo (“How to run a …”)
Blog (future)Article + Author(Person) + BreadcrumbList

Why it matters for AEO: FAQPage/HowTo give answer engines pre-extracted Q→A pairs; Organization

  • sameAs builds the entity graph; SoftwareApplication makes us eligible for tool/product answers.

4. AEO / GEO specifics (the differentiator)

  1. Answer-first copy. Every tool page opens with a 1–2 sentence direct answer to “What is X / how do I do X?” before the UI — the snippet an LLM can lift verbatim.
  2. FAQ blocks on each page (visible HTML and FAQPage JSON-LD) targeting real questions (“Does llms.txt actually help AI visibility?”, “What is a good E-E-A-T score?”).
  3. llms.txt + llms-full.txt at the root (generated by our own tool) — curated map of the site for LLM crawlers.
  4. AI-bot directives in robots.txt: explicitly allow the answer-engine crawlers (many sites accidentally block them; we score this).
  5. Extractable structure: short paragraphs, definition lists, comparison tables, numbered steps — formats LLMs quote cleanly.
  6. Entity & topical authority: a glossary of AEO/GEO terms (AEO, GEO, llms.txt, E-E-A-T, RAG, answer engine, citation) interlinked into a topic cluster; consistent NAP/brand entity.
  7. Citations & comparisons: “AEO vs SEO”, “AEO Toolkit vs ”, “best free AEO tools” — the comparison/listicle queries LLMs synthesize from.
  8. Freshness signals: dateModified in schema; the blogging agent ships regular, deduped content.

5. Content roadmap

  • Pillar: / + a /guide/answer-engine-optimization cornerstone (what/why/how, links to every tool).
  • Tool pages as answer pages: each /tools/* doubles as the ranking page for its keyword (“llms.txt generator”, “E-E-A-T checker”) with intro + FAQ + how-to.
  • Glossary: /glossary/* definition pages (great for AI definitional citations).
  • Comparisons: /compare/aeo-vs-seo, /compare/*.
  • Blog: driven by @advance-labs/blogging (the cron) once LLM keys are set — query-gap-targeted posts.

6. Off-page / authority

  • Dogfood the Backlink Graph to track our own referring domains.
  • Open-source flywheel: the Apache-2.0 repo + the free tools earn natural links/mentions.
  • Submit the Chrome extension (Web Store listing links back), the MCP servers to MCP directories, and the tools to AEO/SEO tool roundups.
  • Digital PR around “we rebuilt SellOnLLM’s tools open-source”.

7. Measurement

SignalSource
Rankings / CTR / impressionsGoogle Search Console (already integrated via /tools/chat + ga-gsc MCP)
Traffic / engagementGA4 (integrated)
AI visibility (are we cited?)our own ai-visibility MCP (Perplexity citation checks) + manual ChatGPT/Claude prompts
Core Web VitalsVercel Speed Insights / CrUX
Index coverage + structured-data validityGSC + Rich Results Test

8. Implementation status

Shipping in this redesign pass: server-rendered design system (the September 2026 brand, see brand/README.md), per-page metadata + canonical, OG images, sitemap.ts, robots.ts (AI bots allowed), /llms.txt, Organization/WebSite/SoftwareApplication

  • per-page FAQPage/HowTo/BreadcrumbList JSON-LD, answer-first copy + FAQ blocks on every page, semantic HTML, internal-link hub, CWV-conscious rendering.

Roadmap: glossary + comparison + guide pages, blog automation (cron), AI-visibility dashboards, Speed Insights, and a programmatic schema audit (run our own auditor against ourselves in CI).