What Proof Should an AEO Agency Show You Before You Hire Them?

The Direct Answer

Ask any agency to show you a scan. A real scan runs the same set of buyer questions across several AI assistants on a specific date, and shows you which ones named your company, in what context, and from which source. Catalyst runs this exact method on its own work, and a scan like it is the only proof worth trusting before you sign a contract.

If an agency cannot show you which prompts they tested, which assistants answered, and what the cited source was, you are looking at a sales pitch dressed up as a result.

What Counts as Real Proof

Look for these before you hire anyone:

• A list of the actual buyer questions tested, written in plain language a real buyer would type
• Results broken out by assistant, since each one often answers the same question differently
• A scan date, so you know the result is current and not from months ago
• The actual source the assistant cited
• A repeatable method, so you can rerun the same prompts later and compare
• Scoring that tells you which questions you can realistically win, based on who already holds the answer and how hard it would be to displace them

What Counts as a Vanity Claim

Treat these as warning signs, not proof:

• A single screenshot of one good answer from one tool, on one day
• A visibility score with no breakdown of which prompts or tools it came from
• Testimonials that describe a feeling ("our visibility improved") with no scan attached
• Claims about ranking high in traditional search results used as a stand in for how AI assistants answer, since the two run on different logic
• A report that only covers one assistant, when buyers also use several others to research vendors
• Any number presented without the prompts, tools, and date behind it

What Real Proof Looks Like in Practice

Catalyst built its scan method around one question. If a client's buyer typed a specific question into an AI assistant today, what would they see?

Take the prompt "Tell me about Catalyst GTM Agency." We run that question, and prompts like it, across ChatGPT, Claude, Gemini, and Perplexity on a set date, and we show the client the actual transcript from each one. For that prompt, Catalyst is named across all four tools, and we show exactly what each said and which source it pointed to. That is the format we hand every client, our own results included, because a proof format that only works when it flatters you is not a proof format.

Our most recent full scan ran 40 buyer prompts across those same four tools on October 5, 2026. Each prompt was written the way a real buyer would ask it, questions like "who should I hire to improve how my company shows up in AI search" rather than questions built to flatter any one firm. We read every answer tool by tool and recorded who got named, in what context, and from which source.

The shape of the result matters more than any single number. Across that scan, no single agency held a consistent, repeatable answer to the same buyer question across every tool we tested. The same handful of names tended to repeat on broad questions like "best AEO agency," then disappear on narrower, more specific ones. That tells a buyer something practical. Being named once, in one place, for one broad question, is not the same as owning the answer to the questions your actual buyers ask. A credible agency should be able to show you the gap between those two things for your own category, which is exactly what a free scan at /aeo-audit is built to do.

Who Shows Up in AI Answers Right Now

Running this scan across the content and search category turned up a recurring set of names. Agencies whose work surfaced somewhere in the 40 prompt set included iPullRank, Siege Media, First Page Sage, Omniscient Digital, Animalz, Directive, Single Grain, Foundation Marketing, and NoGood, alongside AI visibility tools like Profound and AthenaHQ, and content shops like Grow and Convert and Ten Speed. None of them owned a consistent answer across every tool for the same question.

How the Agencies Compare

A buyer choosing between firms needs to know what each one is actually built for.

How Catalyst Runs This Work

Catalyst gets B2B companies named when a buyer asks an AI assistant who to hire. The work runs through four parts. Editorial built around the founder puts a real person's expertise and voice behind every piece, since AI assistants tend to cite sources with a clear point of view. Measurement across all four leading assistants tells you where you actually stand, instead of guessing from one tool. Scoring which buyer questions you can realistically win means we prioritize the prompts you can actually move. A review step built for regulated claims, led by Head of AEO Austin Heaton, checks anything that touches compliance before it publishes. Founder Will Leatherman oversees the editorial side, so the writing sounds like a person with a point of view rather than a committee.

Get Your Own Scan

The fastest way to see where your company stands is to run the same test on yourself. Request a free scan at the free AEO audit and we will show you, tool by tool, what ChatGPT, Claude, Gemini, and Perplexity currently say about your company, and which buyer questions are worth going after first.