July 2026 edition · Free, no gate · Updated monthly

The AEO Playbook

The answer engine optimization playbook, built as numbered plays from 1,237 measured AI citations. Ten chapters; read them in order and leave with your own 90-day plan.

01 What AEO Is, and Why SEO Alone Won't Get You Cited The foundations, briefly, with links to the deep definitions. You build: Your AEO one-liner Read → 02 How AI Engines Choose Their Sources Memory vs live search, and why the four engines barely overlap. Measured on 188 answers. You build: Your engine-reality check Read → 03 Audit: Find Out Where AI Mentions You Today Your baseline and your 10 money prompts. You build: Your baseline scorecard + 10 money prompts Read → 04 Is Your Category Listicle-Led or Vendor-Led? The two-hour measurement that decides where your budget goes. You build: Your category verdict + gap list Read → 05 On-Page AEO: Make Your Pages Answer-Ready Answer-first structure, schema, entity clarity, and the crawler basics. You build: Your ranked on-page fix list Read → 06 Why These Brands Win AI Citations: Teardowns With Data Zapier, Reddit, G2, and Vercel, examined closely, plus verified case studies. You build: Your content plan Read → 07 Off-Page AEO: Earn the Citations That Actually Win The biggest lever: entity records, reviews, directories, Reddit, best-of lists, press. You build: Your 20-surface target list + first 5 pitches Read → 08 Win Each Engine: Data-Backed Profiles What each engine measurably does differently, and the one move per engine. You build: Your per-engine priorities Read → 09 Measure It: Prove AI Search Is Working Citation share, the weekly routine, and honest reporting. You build: Your scorecard + routine Read → 10 Assemble Your 90-Day AEO Plan No new plays: your worksheet entries, turned into a dated week-by-week plan. You build: Your 90-day AEO plan Read →

Filled a worksheet already? See your plan so far →

Where to start
AI never names you. Start with the audit (or the free checker), then entity and reviews first: Play 19 · Play 20
You appear, but rivals win. Classify your category, build the gap list, then work off-page against it: Chapter 4 · Chapter 7
You lead and want to defend. Freshness and measurement, plus the radar so nobody enters your lists unnoticed: Chapter 9 · Play 26
The play index

Every play, numbered.

Around 35 plays are planned across the ten chapters; 35 are live today. Each is a self-contained action card: what to do, exact steps, what it costs in time, what done looks like, and a check you can repeat.

Play 01 Write your AEO one-liner Any Play 02 Run the memory-or-search test on your own category Any Play 03 Pick your primary engine and set an engine budget Any Play 04 Write your 10 money prompts Any Play 05 Run your baseline citation check Any Play 06 Read the answers, not just the score Any Play 07 Set your scorecard and a realistic target Any Play 08 Classify your category as listicle-led or vendor-led Any Play 09 Build your gap list Any Play 10 Rewrite your top pages answer-first Any Play 11 Write your headings as the questions buyers ask Any Play 12 Add the schema types AI reads Any Play 13 Make your brand one clean entity Any Play 14 Fix your internal links Any Play 15 Let the AI crawlers in Any Play 16 Tear down the three pages that win your category Any Play 17 Build an honest comparison page B2B SaaS, Ecommerce Play 18 Add the evidence layer everywhere Any Play 19 Claim your entity on Wikidata, Crunchbase, and LinkedIn Any Play 20 Get your first 10 reviews on G2 and Capterra B2B SaaS, Agency Play 21 Submit to the directories that already cite your competitors Any Play 22 Build your listicle target list and pitch the authors Any Play 23 Show up on Reddit before you need it Any Play 24 Turn journalist requests into earned citations B2B SaaS, Ecommerce, Agency Play 25 Replicate your competitors' most winnable links B2B SaaS, Ecommerce Play 26 Run a listicle radar so you never miss a new list Any Play 27 Win ChatGPT Any Play 28 Win Perplexity Any Play 29 Win Gemini Any Play 30 Win Claude Any Play 31 Win Google AI Overviews Any Play 32 Build your citation scoreboard Any Play 33 Segment AI referrals in your analytics Any Play 34 Set the routine: weekly check, monthly report, quarterly review Any Play 35 Report it honestly Any, Agency
What this is

A manual you run, not an essay you read.

The AEO Playbook is a free, open manual for getting your brand cited by ChatGPT, Perplexity, Gemini, Claude, and Google AI Overviews. Every chapter ends with a worksheet that produces something for your own brand (your money prompts, your gap list, your target list), and the Your plan page assembles your entries into one document you can copy or download.

The playbook's method in six steps, ending in an AI answer that names your brand: audit your visibility across ChatGPT, Perplexity, Claude, and Google; classify your category as listicle-led or vendor-led; fix your pages; ship citable content; earn placements on G2, Reddit, Product Hunt, and Wikidata; and measure citation share weekly per engine, judging the trend rather than single runs.
The six steps of the playbook, and the payoff they aim at: your brand named in the answer, with the sources AI actually cites behind it.

Every tactic is backed by a number from our 1,237-citation study or a named external source, and the one number the whole book hangs on is this:

Bar chart. Share of all 1,237 citations, by source type: Independent listicles 40%, Vendors' own sites 34%, Media & SaaS blogs 15%, Reddit & forums 4%, Review sites (G2 etc.) 3%, YouTube 3%.
Source-type share of 1,237 citations across ChatGPT, Perplexity, Gemini, and Claude, measured 2026-07-10 on 47 buyer questions across 8 industries. Independent best-X listicles led at 40%, ahead of the vendors' own sites at 34%. Full study
Reuse this chart

Free to reuse with attribution. Copy the embed code, or save the image.

Download PNG 
<figure>
  <img src="https://tryprefer.com/figures/citation-2026-07-source-types-hub.png" alt="Bar chart. Share of all 1,237 citations, by source type: Independent listicles 40%, Vendors' own sites 34%, Media & SaaS blogs 15%, Reddit & forums 4%, Review sites (G2 etc.) 3%, YouTube 3%." width="780" height="259" style="max-width:100%;height:auto" />
  <figcaption>Where AI search citations actually go by <a href="https://tryprefer.com/playbooks/aeo-playbook/">Prefer</a></figcaption>
</figure>

Independent lists beat the brands' own websites. That is why this playbook spends so much of its effort on the places other people write about you, and it is first-party data you can check, copy, and cite from the statistics page.

What we don't claim

Two honest limits.

ChatGPT's memory answers

29 of 47 answers in our study came from memory with no live search. Nobody has a proven fast path into a model's memory; the honest answer is consistent third-party coverage over quarters, not weeks.

llms.txt

We publish one, but the evidence says it is hygiene, not a lever: Ahrefs' June 2026 study of 137,210 domains found only 3% of llms.txt files received any request at all, and Google says it does not use the file.

Changelog
Jul 15, 2026
July 2026 edition, v0.3

Index-first hub, brand logos throughout, richer figures, and the Your plan page: worksheet entries now assemble into one document you can copy or download.

Jul 14, 2026
July 2026 edition, v0.1-0.2

Hub plus five chapters shipped: engines (2), category patterns (4), teardowns (6), off-page (7), and measurement (9). Nineteen plays live. Case-study standard: every external claim named, linked, dated.

The AEO playbook, answered.

What is the AEO playbook?

The AEO Playbook is Prefer's open, step-by-step manual for answer engine optimization: getting your brand cited and recommended by ChatGPT, Perplexity, Gemini, Claude, and Google AI Overviews. It is built as an action system, not an essay: around 35 numbered plays with exact steps and verification, one worksheet per chapter, and by the final chapter your entries assemble into your own 90-day AEO plan. Every tactic is backed by a number from our published 1,237-citation study or a named external source.

Is the AEO playbook free? Is it gated?

The full playbook is free and ungated on the web, every chapter, every play, and it will stay that way, because content behind an email gate cannot be read or cited by AI engines and we would not follow our own advice by gating it. A companion pack (a designed PDF, a duplicatable Notion and Sheets checklist, and a prompt library) is coming as an optional email download.

How often is the playbook updated?

Monthly, and the changelog on this page says exactly what changed. The AI engines change behavior often enough that a static guide decays within months, so chapters carry visible last-updated dates and tactics that stop working get corrected in public rather than quietly removed.

What data is the playbook based on?

The spine is the Prefer AI Citation Study: 47 real buyer questions across 8 industries, asked to ChatGPT, Perplexity, Gemini, and Claude with web search on, producing 1,237 logged citations across 713 domains (July 2026). Third-party claims are cited to named sources and labeled as such, and vendor case studies that name no customer are excluded on principle.

By Javed Khatri, Co-founder, Prefer Last updated July 15, 2026

Go deeper

Tools, templates, and directories The appendix: every tool the plays use (free-first), where to get listed, and copy-paste templates. The study behind the playbook Who gets cited in AI search: 47 buyer questions, 1,237 citations, five findings. AI search citation statistics Every number in the playbook, anchor-linked and copy-ready with sources. Free AI visibility checker Your chapter 3 baseline, automated: see whether AI cites you, in about 15 minutes. Have the playbook run for you Prefer Managed executes these plays end to end, reported against your own numbers.

Read it. Run it. Keep the plan.

Start at chapter 1, fill in the worksheets as you go, and your 90-day plan builds itself. Prefer to see where you stand first? Run the audit. The web version is free and stays free.

Start at chapter 1 Or run the audit first