AEO process for getting businesses cited in AI search, learn our exact 5-stage agency workflow to appear in ChatGPT, Perplexity & Gemini in 60–120 days.
You've been comparing approaches. You've read the surface-level explainers and watched the YouTube breakdowns. Now you want to know exactly what a working AEO process looks like, step by step, so you can either run it yourself or evaluate an agency with confidence.
We built this workflow the hard way: first on our own site, then refining it account by account. When we ran it on automatethejourney.com, we found our own site was quietly blocking the AI crawlers and had no llms.txt, the exact gaps that keep a business out of AI answers. Here is the process, stage by stage.
Answer Engine Optimization (AEO) is the practice of structuring your content, authority signals, and citations so that AI systems like ChatGPT, Perplexity, Gemini, and Bing Copilot pull your business into their generated answers.
Traditional SEO earns you a ranked link. AEO earns you a spoken or written mention inside the AI's answer itself. Those are fundamentally different outcomes. A ranked link requires a click. A citation in an AI answer delivers your business name, location, and positioning directly to the reader, no click required.
The practical difference: when someone asks Perplexity "who's the best HVAC company in Scottsdale," your business either appears in that answer or your competitor does.
Businesses with structured FAQ content, consistent off-page citations, and schema markup tend to start appearing in Perplexity answers within 60 to 90 days. Businesses with none of those elements often do not appear at all, even when they rank on page one of Google.
We never skip the audit. Jumping straight into content production without one wastes time and produces the wrong content.
Run these four assessments before touching your website:
Skipping the manual AI query test is the single biggest mistake we see. Teams assume they're invisible in AI answers when they're actually appearing, just for the wrong services or in the wrong geographic area. Always audit before you build.
For example, a clinic might already be cited by Perplexity for a broad term like "anxiety therapy" but be invisible for a higher-margin service like "EMDR therapy." The audit surfaces exactly that kind of gap, so you know where to focus next.
AI systems don't reward volume. They reward clarity. A 600-word page that directly answers one specific question outperforms a 3,000-word generalist page every time in AI citations.
We organize every service site around question clusters, not keyword clusters. The difference matters.
A keyword cluster groups pages around "HVAC repair Scottsdale," "AC repair Scottsdale," and "furnace repair Scottsdale." A question cluster groups pages around "How much does AC repair cost in Scottsdale?", "What causes an AC unit to stop blowing cold air?", and "How do I choose an HVAC company in Scottsdale?"
Question-format content maps directly to how AI systems parse queries. When someone asks Gemini a question, Gemini pulls from sources that answer that exact question, not from sources that merely contain the keyword.
Every service page we build follows this architecture:
Write your direct answer paragraph as if you're answering a friend's text message, one clear sentence, no hedging. AI systems pull the most direct, confident answer they can find. Vague or qualified openings get skipped.
A single well-structured page rarely earns AI citations on its own. AI systems prefer sources they've seen answer related questions across multiple pages. This is topical authority: owning a subject area comprehensively, not just a single keyword.
We run a 90-day content sprint for each client using this sequence:
A focused pillar page (say, "interior painting cost" for a specific city) is often what moves a business from zero AI mentions to showing up in Perplexity answers for that exact query.
Schema markup is structured data you embed in your page code. It tells AI systems, not just Google, exactly what your business does, where you operate, what questions you answer, and what credentials you hold.
Most service businesses have no schema at all. The ones that do have only basic LocalBusiness schema. We install five schema types minimum on every AEO engagement.
We use Google's Rich Results Test and Schema.org's validator to verify every implementation before publishing. Broken schema does more harm than no schema. AI systems that can't parse your structured data ignore it entirely.
Don't install FAQPage schema on a page that doesn't actually display FAQ content visibly on the page. Google and AI systems both check for consistency between your schema and your visible content. Mismatches get flagged and your schema gets discarded.
Content architecture and schema handle what AI systems learn from your website. Off-page citation seeding handles what they learn about you from everywhere else.
AI systems train on the open web. They absorb data from directory listings, review platforms, podcast appearances, PR placements, and third-party mentions. A business mentioned consistently across authoritative external sources gets cited more than a business that only exists on its own website.
We run this sequence for every new AEO client:
Third-party mentions act as votes that AI systems count. One legitimate guest post on a local chamber of commerce website carries more weight than 50 directory listings on unknown aggregators. Quality over quantity applies here even more than it does in traditional link building.
Measuring AEO results requires a different process than checking Google rankings. There's no rank tracker for ChatGPT. You test manually, and you test systematically.
Run this test sequence every two weeks during the first 120 days:
We track this in a simple Google Sheet with one row per query per platform per date. After 60 days, patterns become clear: which platforms cite you most, which query types trigger citations, and which competitors are consistently appearing where you aren't.
Set a recurring 30-minute block every two weeks specifically for manual AI citation testing. The teams that measure consistently are the ones who can connect specific content changes to citation gains, and that feedback loop accelerates the whole process.
Expect first AI citations roughly 45 to 75 days in. Full, consistent presence across the major AI platforms typically takes 90 to 120 days when all five stages run in parallel.
Each stage reinforces the others. Schema without content gives AI systems structure but nothing to say. Content without off-page citations gives AI systems good answers from an unknown source. Citations without schema give AI systems mentions they can't fully parse.
The workflow runs in parallel, not in sequence:
The businesses that win in AI search aren't the ones that publish the most content. They're the ones that answer specific questions more clearly, more consistently, and from more trusted sources than anyone else in their market.
An AEO baseline is a snapshot of where your business stands in AI answers before you change anything: which prompts mention you, which models are strongest and weakest for you, who the AI cites instead of you, and which technical signals (crawler access, schema, llms.txt) are missing. It matters because you cannot prove improvement later if you never nailed down the starting point. The first stretch of any serious AEO engagement is mostly measurement and setup for exactly this reason.
The technical fixes (crawler access, schema, llms.txt) can take effect within days to a few weeks as the engines recrawl you. The harder part, earning consistent third-party citations and brand mentions, builds over months. We treat the first 90 days as foundation and measurement, with the compounding gains arriving after.
The process is industry-agnostic, but it works best where the buying questions are high-intent and the local online competition is weak, which describes most local service businesses. The prompts and competitor sets change per industry; the stages do not.
Yes. You can check your own crawler access, add schema and an llms.txt, claim and complete your Google Business Profile, and ask the AI engines your own buyer questions to see if you are named. A fast way to start is our free AEO audit, which checks the core technical signals for you in about a minute. The deeper measurement, competitor tracking, and citation outreach is where most businesses bring in help.
We load a set of real buyer prompts into an AI-visibility dashboard, capture a baseline across the engines (ChatGPT, Perplexity, Google AI Overviews), and track it against the client's real competitors over time. That turns "are we showing up in AI" from a guess into a number we can move.
We run this exact five-stage workflow for service businesses, local companies, and B2B SaaS teams that are serious about showing up where their buyers are actually asking questions.
If you want us to run your AEO audit, and show you exactly where you stand across ChatGPT, Perplexity, Gemini, and Bing Copilot, book a free strategy call with our team. We'll map your current AI visibility, identify your fastest citation opportunities, and tell you honestly whether this is something you should run in-house or hand off to us.
Get Your Business Cited in AI Search Results Within 120 Days
We'll audit your current AI visibility, identify exactly why competitors are being cited instead of you, and map out your custom 5-stage AEO roadmap, no fluff, no guesswork.
Written by Tim Hershberger, founder of Automate the Journey. Tim has built 500+ marketing automation systems for service businesses since 2009. Book a free strategy call to see how we can help.
Everything in this guide runs on GoHighLevel. Try it free for 30 days and see why we chose it.
No credit card required · Cancel anytime