The 90-Day AEO Playbook: From Zero to 40% MentionShare

Why 90 days?

The AI citation cycle works differently from SEO. Content changes can show up in Perplexity within 48 hours. ChatGPT incorporates new training data in waves. Gemini refreshes more frequently than most brands realize.

90 days gives you enough time to publish, measure, iterate, and measure again — without the multi-year timeframe of traditional SEO.

Days 1–7: Baseline measurement

Before you change anything, measure where you stand.

Step 1: Set up TrueCite with your business and 10–15 buyer-intent prompts. Prompts should mirror what your ideal customer would ask when researching vendors: "What is the best [category] tool for [use case]?"

Step 2: Run your first scan across ChatGPT, Perplexity, and Gemini. Record your MentionShare score for each engine.

Step 3: Look at who is getting cited. Which competitors are in the answers? What sources are being cited?

Days 8–30: Technical fixes

These are the quickest wins and should be done before any content changes.

Organization schema (Days 8–10):

Add Organization JSON-LD to your homepage with your brand name, website, logo, description, and sameAs links to G2, Capterra, LinkedIn, Twitter/X.

llms.txt file (Days 10–12):

Create /llms.txt with your company context, key page links, and external citation links.

FAQ schema audit (Days 12–20):

Do you have any FAQPage schema on your site? If not, pick 3–5 existing pages with question-like headings and add schema markup.

Review platform push (Days 15–30):

Email your 20 happiest customers asking for G2 and Capterra reviews. A targeted ask typically yields 5–10 reviews from a batch of 20 emails. This alone is often enough to trigger Perplexity citations.

Days 31–60: Content creation

Now you build the content that AI engines will cite for the next 12+ months.

Core FAQ page (Days 31–40):

Create a /faq page targeting 15–20 buyer-intent questions. Structure: question in `<h2>`, direct answer in the first sentence, 1–2 supporting sentences. Add FAQPage JSON-LD schema. TrueCite's [Fix Generator](/features/fix-generator) creates these automatically.

Answer paragraphs (Days 40–50):

For each prompt where AI currently skips your brand, create a short answer paragraph (150–250 words) that directly answers the question and mentions your brand naturally.

Comparison content (Days 50–60):

Create one "[Your Brand] vs [Main Competitor]" page for your top 2–3 competitors. These pages are heavily cited by Perplexity for comparison queries.

Days 61–90: Measurement and iteration

Run weekly scans and track [MentionShare](/features/mentionshare) for each engine.

Identify remaining gaps. Use [Competitor Intelligence](/features/competitor-intelligence) to see which prompts your competitors are winning that you are not. Use the [AI SEO Audit](/features/ai-seo-audit) to surface any pages still missing schema.

Add more AI engines. If you started with ChatGPT, Perplexity, and Gemini, add Claude and Grok. Different engines have different citation patterns.

What 40% looks like

Forty percent MentionShare means AI recommends your brand in 2 out of every 5 answers to buyer-intent queries. For most B2B categories, this puts you among the top 2–3 recommended brands.

The brands that hit 40% in 90 days typically have:

• ▸An FAQ page with 15+ questions and FAQPage schema
• ▸Organization schema with 4+ sameAs links
• ▸20+ G2 reviews
• ▸llms.txt file
• ▸2–3 targeted answer paragraphs per high-priority prompt

None of this is technically complex. All of it is achievable in one focused sprint. [Start your AEO journey](/sign-up) — run your baseline scan and have your first fix published in 48 hours.