AI search
AI search optimisation
AI search optimisation is the work of making a business readable and quotable by the systems people now ask instead of searching: ChatGPT, Google's AI Overviews, Gemini, Perplexity. It is good local SEO made machine-readable. A new front door, the same lock. Built for Australian trade and home-services businesses.
Why most businesses are invisible to AI answers
An assistant asked who should I call in Kambah does not return ten blue links. It names two or three businesses. Being absent from that answer is not a ranking problem you can see, which is exactly why almost nobody has checked.
Nothing worth quoting
Assistants lift sentences. A site made of headline fragments and stock reassurance gives them nothing liftable, so the answer gets assembled out of a directory listing instead of out of your own words.
Facts that contradict each other
Name, address, phone and hours disagreeing across your site, your profile and the directories. A model reading conflicting facts about you tends to resolve the conflict by naming somebody else.
No answers on the site
What people ask an assistant are questions, not keywords. A site with no plain answers to the real ones has given the retrieval step nothing to retrieve.
Never measured
Almost no trade business has ever asked an assistant who it recommends in their own suburb and written down what came back. Without a baseline there is nothing to improve against.
What tyso does about it
The same work that makes a business rank, done so machines can read it, plus a measurement habit. It is included in every engagement rather than sold as a mystery add-on.
Answers, written plainly
Real questions with real answers, put where they can be lifted whole: the opening lines of a page, not buried under a scroll of atmosphere. Assistants pull hardest from the top.
Facts made to agree
One name, one number, one address, one set of hours, matching everywhere a machine might read them. Boring work, and the most common single reason a business gets skipped over.
Structured data that matches the page
Schema markup that says the same thing the visible copy says. Machine-readable is not a separate story told to robots: it is the same story, in a format they can parse.
A baseline, then a re-test
The questions get asked on day one and the answers written down, then asked again later. Nobody controls a model's output, so the honest version of this work is measurement rather than a promise.
What can be checked
This is the easiest part of the studio to bluff and the easiest to verify, so here is the measured version. tyso is a one-person Canberra studio, launched in 2026.
Quoted by Google's AI Overviews
ANT Painting, whose search presence tyso runs, was cited by Google's AI Overviews on 7 of 22 painting questions we measured in August 2026.
The probe runs live in the audit
The free audit reads your site, measures your mobile speed with Google PageSpeed, and asks an AI assistant who it recommends in your suburb. That last check is this service, run on you, for nothing.
The old signals still count
Sophisticated Manes went from 3 Google reviews to 10 in its first month live. Assistants read the same trust signals people do, which is why this work never leaves local SEO behind.
The working, written up
How it starts
With the audit, because the AI visibility probe inside it is this service running live. It asks an assistant who it recommends in your suburb and shows you the answer, name in it or not.
- Give it your business, your trade and your suburb.
- It puts the question to an assistant and reports what came back.
- That answer is your baseline, whether we end up working together or not.
Questions about AI search
What is AI search optimisation?
The work of making a business readable and quotable by AI assistants and AI Overviews, so the answer they give has a decent chance of being you. In practice: good local SEO, plus plain answers, consistent facts, matching structured data, and measurement.
Is generative engine optimisation the same thing?
Yes. Generative engine optimisation, answer engine optimisation and AI search optimisation all describe the same job: being the source an AI answer gets built from. The names came out of different corners of the industry and none of them has won yet.
How is AI SEO different from ordinary SEO?
Less different than the name suggests. The signals overlap heavily, and a business that ranks well locally is usually one an assistant can cite. What changes is format: machines need liftable answers, facts that agree, and structured data, near the top of the page.
Can you guarantee ChatGPT will recommend me?
No, and be careful with anyone who does. Nobody controls a model's output. What can be done is optimise the signals it reads, measure what it says about you on day one, then measure again later and show you the difference either way.
How do you measure whether it worked?
By asking. A set of the questions your customers would really ask, put to the assistants, with the answers written down and dated. Then the same questions again after the work. A baseline and a re-test, not a dashboard number that flatters everyone.
Ask it what it says about you.
The free audit puts the question to an AI assistant and shows you what came back, name in it or not. No email needed to see the score.
Rather talk it through? Book a call.