GEO audit: find out why AI answers skip you

What a GEO audit is, and what it is not

A traditional SEO audit crawls your site and reports on technical health, keywords and backlinks. A GEO audit starts from the other end. It begins with the questions a buyer types into an AI assistant, reads the answers that come back, and works backwards to explain why you were or were not named.

That difference matters because the inputs differ. An AI answer is assembled from a small set of pages the assistant trusts for a topic. Your homepage is one possible input. Competitors' comparison pages, documentation, review grids, roundup articles and forum threads are others, and in most categories they carry more weight than anything you publish yourself.

A GEO audit is not a guarantee of placement, and anyone selling one should say so. There is no paid slot in organic AI answers. What an audit gives you is evidence: where you stand today, which sources are shaping the answer, and which changes are most likely to move it.

What our own data shows

We ran unbranded buyer questions ("best [category]", "[competitor] alternatives", "X vs Y") through AI search for 50 software brands. These are the only numbers we quote anywhere on this site, and we quote them as they are.

• 39 of 50 brands (78%) were never named in any answer to their own category's buyer questions.
• All 30 brands that had launched in the previous month scored 0.
• Being known is not enough: an established e-signature tool scored 52/100 and an AI voice tool scored 37/100.
• In 33 of 50 cases the most cited source was a competitor's own website: its "vs" pages, docs and help centre.
• Review sites were the most cited source for only 6 of 50 brands.

What a proper GEO audit covers

1. 01Prompt set. 10 to 20 unbranded questions real buyers ask in your category: "best X for Y", "X alternatives", "X vs Y", "cheapest X for a small team", "tools like X". Branded questions are excluded because they measure recall, not discovery.
2. 02Live answers. Each prompt is answered with web search turned on, because that is what buyers see. Answers vary between runs, so the audit samples rather than relying on one screenshot.
3. 03Mention analysis. For each answer: were you named, in what position, and how were you described? A mention that calls you "a budget option for freelancers" when you sell to enterprises is a problem of its own.
4. 04Competitor share of voice. Which brands take the slots you want, and how often. This is usually the most persuasive page of the report for a leadership team.
5. 05Cited sources. Which domains and pages the answers lean on, ranked by frequency. This becomes the work plan.
6. 06On-site readiness. Is your category stated in one plain sentence in the title, H1 and meta description? Do you have comparison pages, public pricing, crawlable docs, structured data and an llms.txt?
7. 07Fix plan. A prioritised list ranked by impact and effort, with the first three actions specific enough to start the same day.

How to run a basic GEO audit yourself in an afternoon

You can do a rough version without any tool. It will be noisier than a structured audit, but it is honest and free.

1. 01Write down your category in the words a buyer would use, not your tagline. "Invoicing software for freelancers", not "the future of getting paid".
2. 02Write 10 questions a buyer would ask without knowing you exist. Mix "best", "alternatives to [market leader]", "[leader] vs [runner-up]" and one with a constraint such as budget or team size.
3. 03Ask each question in two or three AI assistants with web search on, in a logged-out or fresh session. Record every brand named, in order.
4. 04Note every source the answers cite or link. Count how often each domain appears across all answers.
5. 05Mark where you appear. Most brands will find they appear rarely or not at all; in our data that was 78% of brands.
6. 06Open the top five cited pages. Ask of each: are we mentioned, and if not, is there a realistic way to be? An author to email, a review profile to complete, a thread to answer honestly?

Reading the results

• Not named anywhere: start with category clarity and one or two sources that appear in many answers. New brands nearly always land here; every brand in our set launched in the previous month scored 0.
• Named for some prompts, missing on comparisons: competitors' "vs" pages are probably writing your story. Publish your own honest comparisons.
• Named but described wrongly: fix the plain-language description on your site and on your directory and review profiles so every source repeats the same sentence.
• Named often: defend it. Keep comparison pages current, watch for competitor overtakes and re-check monthly.

The fixes that come up most often

• One plain category sentence, used identically in the title tag, H1, meta description, directory listings and social bios.
• Honest comparison and alternatives pages. In 33 of 50 brands we checked, the most cited source was a competitor's own site, so the brands that publish these pages tend to frame the answer.
• Public, specific pricing. Answers prefer products whose price they can state.
• Ungated docs and help-centre pages. Gated PDFs cannot be read or cited.
• Presence in the roundups the answers already cite. A single mention in a frequently cited article can be worth more than many new blog posts.
• FAQPage and Organization structured data plus an llms.txt that summarises who you are and links your key pages.

Free check or done-for-you audit

The free ShownInAI check runs 10 unbranded buyer prompts for your domain and returns a 0 to 100 score, the competitors named instead, the cited sources and the first fix. It takes a couple of minutes and needs no account.

The done-for-you GEO audit is $299, one-time. It covers 20 buyer queries, a competitor comparison, the cited sources and a prioritised 30-day fix plan, delivered within 48 hours, with an optional 30-minute walkthrough. Agencies can white-label it and resell it to clients.