AI SEO audit: what it checks and what it costs (2026)

An AI SEO audit checks whether Google and the AI answer engines can crawl your site, read it and name you for your buyers' questions. Then it ranks the fixes. The engines that matter are ChatGPT, Gemini, Perplexity, Claude, Copilot, and Google's AI Overviews and AI Mode. Prices run from free automated checkers to about US$5,000 for a done-for-you audit in the US, and A$590 to A$2,500 in Australia. Same job, many names. An AI visibility audit, a GEO audit and an AEO audit all describe it.

I'm Brenden House, and I run Searchmaxxed, an SEO and AI SEO agency. On 27 October 2026 (AEST) I read every audit price and scope in this article on the provider's own page, and dated each one.

In summary

An AI SEO audit costs nothing as an automated page check, and US$1,500 to US$5,000 or A$590 to A$2,500 done for you.

• A full audit covers six layers, from crawler access to the outside sources AI engines cite.
• Free tools check pages. They can't see your buyer questions, your competitors, the sites AI answers quote or your GA4 data.
• Price follows scope: questions tested, engines checked, runs per question, and whether a person ranks the fixes.
• Judge the output, not the score.

What is an AI SEO audit?

An AI SEO audit reviews how your website and brand appear in Google and in AI-generated answers, then ranks the fixes. It takes a classic SEO audit and adds two questions. Can AI crawlers read us? And who do the AI engines name when our buyers ask?

The names are messy:

• AI SEO audit is the umbrella term, checking Google rankings and AI answers together.
• AI visibility audit weights the answers: who gets named, cited or left out.
• GEO audit (generative engine optimisation) asks whether engines like ChatGPT name and cite you when they write an answer.
• AEO audit (answer engine optimisation) checks whether your pages hold answers an engine can lift whole, including AI Overviews and featured snippets.
• A citation is a link or source an AI answer shows beside its text.
• A prompt is a question put to an AI engine, phrased the way a real buyer would ask it.

Here's how it differs from the SEO audit most marketing teams already know.

Google's own guidance explains the overlap. Its bar for AI Overviews and AI Mode is the normal one. A page must be indexed and "eligible to be shown in Google Search with a snippet". Then the blunt part: "There are no additional technical requirements." (Google Search Central)

So a good AI SEO audit starts with classic SEO. Then it adds what Google's checklist leaves out, because ChatGPT, Perplexity and Claude run their own crawlers and choose their own sources.

At Searchmaxxed we run both halves as one audit, because a fix on one side usually helps the other. Our SEO audit service lists what you get.

What does an AI SEO audit check?

A complete AI SEO audit checks six layers: crawler access, rendering and indexing, entities and schema, answer-ready content, live answer tests and outside sources. Most free tools stop after the third or fourth.

1. Can AI crawlers reach the site?

The audit reads robots.txt for each bot by name, then checks the CDN and firewall, which catches blocks robots.txt never shows. Google puts "ensuring that crawling is allowed in robots.txt, and by any CDN or hosting infrastructure" first among its basics (Google Search Central).

OpenAI splits its bots by job. Its crawler documentation says a site "can allow OAI-SearchBot in order to appear in search results while disallowing GPTBot" for model training. So a security team that blocks every OpenAI agent at the firewall can switch off ChatGPT search visibility by accident. On enterprise sites, that rule usually belongs to someone outside marketing.

2. Does the content render without JavaScript?

Plenty of SaaS and product sites build their pages in the browser with frameworks like React or Angular. If the words only appear after scripts run, some AI crawlers see an empty shell. The audit compares raw HTML with the rendered page on every main template and flags any that hides its main text.

3. Do the entities and schema agree?

An entity is a thing a search engine can identify: your company, products, founders or locations. The audit checks that your schema names each one exactly as your pages, LinkedIn page, Crunchbase entry and review profiles do. Google wants structured data that "matches the visible text on the page". Mismatched schema is a fault.

4. Is the content built to be quoted?

AI answers lift short passages that make sense on their own. So the audit looks for question-led headings with the answer up top, plain definitions, and tables wherever a buyer compares options. It also checks author names, dates and sources, the signals behind what Google calls E-E-A-T: experience, expertise, authoritativeness and trustworthiness. Then it lists the pages you don't have. For a B2B company that's usually comparison, alternatives, integration and pricing pages, which engines go looking for whenever a buyer asks for a shortlist.

5. What do the engines say today?

Free tools skip this layer entirely. The audit puts a fixed set of buyer questions to each engine several times. It records who gets named, who gets cited and what the answer says about you. Answers change between runs. One test on one day proves little, so a sound audit repeats each question and reports a range.

6. Which outside sources decide the answer?

Each engine leans on a short list of trusted sites for your category, and the audit maps that list for your buyer questions. Are you on it?

What do free AI SEO audit tools miss?

Free AI SEO audit tools score a page's technical readiness. They can't test your buyer questions, map the sources AI answers cite or check your analytics, and those decide whether you get recommended.

Our own testing shows how much weight the sources carry. We put 300 Australian agency-buyer questions to ChatGPT three times each, with search on. It pulled 17,870 search results into those 900 answers and kept 4,389 as sources. Being found is the easy part. About three in four results it reads never make the answer.

And the sources it keeps bunch together. Across 900 ChatGPT answers about Australian agencies, a single third-party "best agencies" site was cited in 168 of them. The next three were a second listicle, Semrush's agency directory and Clutch. If you want to be recommended, the job is getting onto the handful of pages the model already trusts.

The engines differ too. Perplexity cited a review directory (Semrush Agencies, Clutch, DesignRush and the like) in 36 of 44 agency answers. ChatGPT did it in 20 and Claude in 15. One agency's Clutch profile had six reviews and was cited 20 times.

A perfect page score fixes none of that. An audit that skips layers 5 and 6 can tell you the site is readable, but not why a competitor gets named in your place.

Free tools still earn their spot. Run one before you pay anyone. If it finds blocked crawlers or empty HTML, fix those first, because every later finding depends on them. If you'd rather track answers yourself, our guide to AI visibility tools compares the software.

How much does an AI SEO audit cost?

A done-for-you AI SEO audit costs US$1,500 to US$5,000 from US providers that publish prices, and A$590 to A$2,500 in Australia. Below that sit free checkers and self-serve audits at US$99 to US$500.

This is my own selection of providers that publish an audit price. I run Searchmaxxed, so it's listed first. Read it as a price map, not a ranking.

Each figure comes from the provider's own page, read 27 October 2026 (AEST). US providers show $ without naming a currency, and every figure is an advertised price rather than a market average.

Guides on page one quote wider ranges. Avante's own cost guide says audits "range from $0 (DIY) to $25,000+ (enterprise)". Pierview's pricing guide puts "a standalone audit" at "$1,500 to $5,000" (Pierview).

What makes one audit cost more than another?

Five things move the price:

1. Questions tested. Ten prompts is an afternoon; 300 prompts, run three times each across several engines and two markets, is a week of work.
2. Engines and runs. Each extra engine and each repeat run adds hours.
3. Site size. A 40-page marketing site needs one crawl, while a 40,000-page site with a dozen templates needs a crawl plan and a sampling method.
4. Markets. Australia and the US cite different sources, so each market needs its own questions, its own runs and its own source map.
5. Who ranks the fixes. A tool sorts by severity. A person sorts by what each fix should earn.

Avante makes a fair point about doing it yourself: "Running a real audit takes 30-60 hours the first time". For a stretched growth team, those hours are the real price.

For the cost of the ongoing work an audit usually leads to, see how much AI SEO costs.

What should an AI SEO audit deliver?

A useful AI SEO audit hands you today's answers, faults tied to what they cost, fixes ranked by value and a dated plan. A score alone is just a reading.

Hold any proposal against this checklist.

That analytics row gets skipped more than any other, and it's where we find some of the worst surprises.

Here's one from an enterprise B2B client's account. A client's GA4 showed AI-assistant visits jump from 354 to 6,050 in a month. Almost none of it was real. About 5,700 came in three one-hour bursts: same 1920x1080 desktop Chrome, all new users, no referrer, about 20 seconds each, landing on two pages. Real AI traffic was 385 sessions against 354 the month before, roughly flat.

And one from the conversion setup. A client's four enquiry types never counted as conversions. Their event names ran 41 to 50 characters; GA4 cuts at 40. The 40-character "career" event counted every time. Nobody had noticed, because the events still showed up.

An AI visibility score would show neither fault, yet both would wreck the numbers every fix gets judged on. That's why we check measurement before recommending a single change.

How long does an AI SEO audit take?

A done-for-you AI SEO audit takes three to five business days from the providers that publish a turnaround. A free checker returns a page score in under a minute.

AI Search Optimization Agency quotes three business days for its Starter audit and five for Pro. Digital Marketing Tribe delivers within five. Searchmaxxed's AI SEO Audit also takes five business days.

Big sites change the maths. When an audit has to cover every template across several markets, set the scope first and ask for the turnaround after.

A worked example: an AI SEO audit for a B2B SaaS company

Picture a B2B SaaS company selling workforce scheduling software in Australia and the US. Its head of growth wants to know why ChatGPT keeps naming two competitors when buyers ask for scheduling software. This is an illustration of how the audit runs, not a client result.

Day 1: data and questions. We connect Search Console and GA4 and test the conversion events first, then write the buyer questions with the team. Think "best employee scheduling software for hospitals", "[competitor] alternatives for multi-site retail" or "how much does workforce management software cost".

Days 2 and 3: answers and sources. Each question goes into ChatGPT, Gemini, Perplexity, Claude, Copilot and Google's AI Overviews, more than once, in both markets. We log every name and every source. Then we read the competitor pages and listicles behind those answers.

Day 4: the site. The crawl checks robots.txt and the firewall for each AI bot, then compares raw and rendered HTML on the product, pricing and integration templates. Schema gets checked against the visible text. Finally we list the pages the answers want and the site lacks, such as alternatives pages or an Australian pricing page.

Day 5: the plan. Every fix gets a value, an owner and a place in the order. Unblocking a crawler might come first. It's cheap, and it gates everything else. A comparison page for the most-asked question might come second, and a push onto the two review sites the engines cite most might come third.

The company walks away without a score. Instead it knows why the competitors get named, and what to ship in the next 90 days to change that.

How does Searchmaxxed run an AI SEO audit?

Searchmaxxed's AI SEO Audit checks your Google rankings and AI answers together in five business days, then ranks every fix by what it's worth. You get one document in four parts.

1. Where buyers search without finding you. The searches that bring revenue, with your Google rank and AI answer presence beside each, and who wins them now.
2. What the site is doing to itself. Crawl, index, rendering, schema and internal-link faults, each tied to the searches it touches.
3. Every fix ranked by what it's worth. A fault on a page nobody searches for drops down the list, while one on a money page moves up.
4. The plan. Fixes in ship order, each naming the page, the change and who does it.

Then we walk you through it live. Your team can run the plan, or we can.

Here's how we do it. Our AI agents connect to your Search Console and GA4, read the Google results and scrape the competitor pages that win. They also put your buyer questions to ChatGPT, Gemini, Perplexity and AI Overviews, and our team reads the evidence and ranks the fixes. If you go ahead, the Managed Search Loop ships the plan weekly on a website you own. The Off-Page Source Layer works on the outside sources.

The audit is credited to the build or first month when you start within 14 days, and either way the plan is yours to keep. Searchmaxxed's work has generated more than A$10 million in revenue through search for its clients.

Is Searchmaxxed's AI SEO Audit right for you?

Searchmaxxed's AI SEO Audit suits marketing and growth leads at enterprise, B2B SaaS, venture-backed, manufacturing and public-sector companies who want a plan they can ship. Want a score and nothing more? A free tool will do.

After the audit, the work runs inside one of our managed packages. The Growth System ships 30 page releases a month. Agentic Search Engineering ships 50, adding digital PR, link campaigns and AI answer tracking across ChatGPT, Gemini, Perplexity and AI Overviews. Custom & Enterprise covers several markets, dedicated engineering, custom reporting and first-party research.

Frequently asked questions

What should you do next?

Write down five questions your best buyers would ask ChatGPT about your category. Ask each engine, and note who it names. If it isn't you, and you want to know why and what to fix first, let's talk pricing.