There’s a family right now in Auckland planning a September trip to Queensland. They’re not opening twelve tabs plus trivago to compare prices, they’re typing one sentence into ChatGPT so they get back a shortlist of places to stay, eat and do. Some Caloundra businesses will make the list and they may not even be the best businesses in town… they’re simply the ones a machine could read.
That’s the shift, it’s already happened and even faster than almost anyone planned for.
Australians are slow to it but our visitors aren’t.
Here’s where it gets interesting for a region like ours – The Australian Holiday Barometer, reported in April that only 12% of Australians have used AI to help plan or book a holiday, compared with 22% of Americans. That tracks because Australians are cautious by nature; we fact-check and then ask a mate who’s already been.
If you’re reading this thinking it’s a problem for accommodation and tour operators, have another look at what visitors are doing; The physio treating a holiday injury, the bakery, the hairdresser, the mechanic, the clothes store, the bloke who fixes boat trailers in January. Plenty of businesses in this region live off the visitor economy without ever calling themselves tourism operators and the same AI machine decides which local business gets named when a Caloundra resident asks for a recommendation.
This isn’t a tourism only phenomenon. There’s a genuine collective effect here, a destination gets recommended by AI when enough of its individual businesses are legible and easy for an AI to describe with confidence that this is a great place to visit.
In the United States, 45% of consumers had used an AI tool to find a local business recommendation in the last 12 months compared to 6% the year before versus Google’s share of online searches for local businesses dropping from 83% to 71% according to BrightLocal’s 2026 Survey. AI is now the third most-used discovery channel for local businesses, sitting behind only Google and Facebook, and ahead of Tripadvisor and other review channels.
Google has already switched on agentic bookings for restaurant tables, event tickets and beauty and wellness appointments in the United States where you describe what you want, and it finds availability and books it through partners like OpenTable and Ticketmaster. Flights and hotels were recently announced to be coming soon. ChatGPT has been doing something similar since late 2025, where people compare real flights and hotels inside a conversation before being shown links to click and book direct.
There’s no confirmed Australian release date but we know it’s definitely coming and we know these same patterns will repeat here because it always does. The lag between the American announcement and the Australian rollout is your preparation window. It isn’t long, and it’s almost certainly the last cheap win you’ll get because when it arrives, the businesses that can be shortlisted by a machine will be, and the ones that can’t will simply sit outside the AI conversation.
What “agentic” actually means, minus the jargon
You’ll hear the word “agentic” a lot this year so let’s strip the jargon out and talk about the one thing it actually means: the AI stops suggesting and starts doing.
AI machines read three things first about your business
- Your own website – AI tools don’t browse a website the way a person does. They look for unambiguous answers to learn and give the answer back for ‘what is this business’, ‘where exactly does it operate’, ‘who is it for’, ‘what does it cost’, ‘is it open’, etc. If that information only exists inside a nicely designed banner image or a paragraph of brand poetry, it may as well not exist.
- Your listings – Your Google Business Profile is now less a directory listing and more the official script an AI reads aloud about you. Wrong hours, missing services, no recent photos and thin categories all reduce a machine’s confidence in naming you.
- What everyone else says about you – This is the one business owners underestimate. A lot of information that AI says about your business doesn’t come from your website at all. It comes from reviews, directory listings, chamber and council pages, local news and industry association mentions, cross-referenced until the machine feels confident enough to say your name in the room.
When AI gets you wrong, the customer blames you
Wrong opening hours, offers that ended last year, a tour that no longer runs and booking links that go nowhere. Consider what that actually looks like at your front door when someone drives from the Hinterland or from Brisbane on a Tuesday because an AI told them you were open, and you’re not. They don’t blame the AI, they blame you, and then they tell people (anywhere they can). Stale data used to mean a missed opportunity bur now it manufactures bad experiences and puts your name on them.
The upside nobody’s talking about
Most coverage of this treats AI as a threat to your website traffic. Fair enough, it is. But look at what happens to the traffic that does arrive. Someone who clicks through from an AI answer isn’t just browsing because they’re bored. They’ve already been given a recommendation they trust, and they’re checking you out with intent. Consumers are trusting AI’s with advice and recommendations so we need to flip the narrative and have AI trust our business information enough to get on those recommendation lists and in front of customers.
A 20-minute test you can run without any tech skill
Open ChatGPT, then Google’s AI Mode, then Gemini, and ask each one a question your actual customer would ask like ‘Where should I stay in Caloundra with kids and a dog,’ ‘Best breakfast near Kings Beach,’ ‘Who does emergency plumbing on the Sunshine Coast.’
Then check three things.
- Are you named at all? If not, take note on who is because that’s your real competitive set in 2026, and it often looks nothing like your Google rankings.
- Is what’s said about you correct and current? Old hours, an old address, a service you stopped offering, all of it points to inconsistent information sitting somewhere out on the web.
- Who’s telling your story on your behalf, and is it the version you’d choose?
Right-click your own homepage, choose “View Page Source”, and search that code for a sentence about your business content. If it isn’t in there, or if you see wrong information leftover from cheap website page builders then AI crawlers will never see it or they’ll pick up on wrong information and will lose trust in your business.
Priority list to fix (in order) to get your business AI ready
- Google Business Profile – correct hours including public holidays, complete services and categories, real photos added regularly, and a booking or contact link that works.
- Name your images properly and write real alt text. “caloundra-beachfront-apartment-balcony.jpg” tells a machine something. “IMG_4823.jpg” tells it nothing, and worse still it tells AI that you’re business is lazy.
- Get Schema (aka structured data) onto your website now. Your business type and location, your individual services, your articles, and links out to your verified profiles. This is the highest-leverage technical fix available and it usually doesn’t touch your design at all.
- ATDW next, if you’re eligible. Check whether your listing is live or expired, update it, and diarise a refresh every season rather than every year.
- Make your key pages answer real questions in the first two sentences, using the words your customers use rather than internal jargon.
- Match your details word for word everywhere. Name, address, phone, service area and description, identical across your site, Google, directories and social profiles. Inconsistency reads as low confidence, and a machine will pick the competitor whose details all line up.
- Ask for reviews, consistently and out loud, and respond to all of them properly. Recency counts as much as rating.
- Check your content is in your page code, it sounds nerdy but it’s as easy as hitting Ctrl-F in a word doc.
- Repeat the above once a quarter.
The time is now
You don’t need to learn AI and you don’t need to buy any new tech. You just need your information to be correct, consistent and readable by a machine, in the handful of places that machines actually look. That’s it, and that’s precisely why so few of your competitors have done it yet so now’s the chance to get in front of the curve and inside the AI conversations.