AI Search Is Changing How Customers Find You. Here’s What Australian SMBs Should Actually Do About It.

This is the second article in a three-part series on what AI search means for Australian small businesses. Part 1 covered Google’s official guidance on AI search and why most “GEO” services are unnecessary. This piece covers the genuine shifts happening in user behaviour. Part 3 translates Google’s I/O 2026 announcements into a practical SMB action list.

A few weeks ago, I wrote a piece arguing that most of what’s being sold as “GEO” (Generative Engine Optimisation) is unnecessary, and that Google had just confirmed it. You can read it here. I stand by every word.

But the responses I got were interesting. A few sharp readers pushed back with a fair point: “Sure, Google says GEO is just SEO, for Google. But what about ChatGPT? Perplexity? Copilot? They don’t all crawl, rank, and cite the same way. Doesn’t that change things?”

They’re right. And I want to address that head-on, because the worst takeaway from my last article would be “nothing is changing, carry on as you were.” That’s not what I said, and it’s not what I think.

The timing on this matters. Two weeks ago at Google I/O 2026, Google announced that AI Mode has passed one billion monthly users, with AI Mode queries more than doubling every quarter since launch. They also confirmed that they’re merging AI Overviews and AI Mode into a single, seamless AI Search experience, live worldwide. The way customers find businesses on Google is being actively redesigned, in real time, at a pace most SMBs haven’t fully appreciated.

Here’s the more accurate framing: the fundamentals of SEO still matter most, but the presentation layer of search is evolving rapidly, and there are real changes in user behaviour that Australian SMBs need to plan for. The foundations stay the same. The layers on top are shifting fast.

This piece is about those layers. What’s genuinely changing, what it means for your business, and what to actually do about it.

First, let me steelman the other side

In my last piece, I cited Google’s official AI optimisation guide, which is pretty blunt: optimising for generative AI search is still SEO. That’s accurate for Google’s own surfaces. AI Overviews and AI Mode are built on Google’s core ranking systems, so if you rank well in regular Google, you’re already in the pool the AI draws from.

But here’s the legitimate counter-argument, and it’s worth taking seriously.

Google’s guidance covers Google’s surfaces. It doesn’t cover how ChatGPT crawls the web (it uses OAI-SearchBot and partners with Bing for its search results). It doesn’t cover how Perplexity selects which sources to cite (its own crawler, its own ranking logic, heavy weighting toward recency and source diversity). It doesn’t cover how Microsoft Copilot surfaces brands within Bing’s index. And it doesn’t fully address the deeper shift underneath all of this: customers are increasingly asking AI tools questions they used to type into Google, and the way those answers are presented is fundamentally different from a list of blue links.

There’s also a fair point about Google itself. They published guidance saying GEO is “just SEO”, and then a month earlier, they were recruiting senior staff to manage the Generative Engine Optimisation ecosystem as a strategic channel. The word “GEO” appears in the job description. So the company telling you GEO isn’t a distinct discipline is internally treating it as exactly that.

You can read that contrast charitably (Google is keeping things simple for site owners while doing more sophisticated work internally) or cynically (Google is downplaying GEO because acknowledging it as a separate discipline would weaken Google’s grip on the search ecosystem). Probably a bit of both.

The honest position is this: good SEO covers most of what works in AI search. But “most” isn’t “all.” There’s a thin layer of additional thinking that genuinely matters, and SMBs who ignore it will lose ground over time to those who don’t.

That’s the position I’m working from for the rest of this article.

What’s actually changing and what to do about each

There are five real shifts happening in how customers find businesses through AI. None of them requires you to throw out your existing SEO strategy. All of them require some adjustment.

Shift 1: Click-through rates are dropping for informational searches

What’s happening: when someone Googles “how long does grout take to dry” or “can I claim my home office on tax,” they increasingly get a complete answer in an AI Overview at the top of the page. They never click through to anyone’s website.

This is about to accelerate. At I/O 2026, Google announced it’s merging AI Overviews and AI Mode into a single, seamless AI Search experience, with the question, answer, and follow-up conversation all happening in a single flow. There’s no longer a meaningful distinction between “regular Google” and “AI Google”; it’s becoming one experience, with AI answers at its centre.

For SMBs who built traffic on educational content designed to answer common customer questions, this is a real shift. Informational traffic is falling. The blog posts that used to draw visitors who then converted are pulling fewer visitors.

What to do about it:

Stop measuring content success purely by traffic. Start measuring it by whether your business is cited or referenced in the AI answers your customers see. If you’re a Brisbane accountant and someone asks, “Do I need to register for GST as a sole trader?” the win isn’t a click, it’s being the firm Google’s AI references when answering. That citation is the new equivalent of a top-three ranking.

Shift your content strategy accordingly: less generic information that AI can repackage without crediting you, more genuinely expert content tied to your brand, your case studies, your specific experience. Information that requires attribution is information AI has to reference rather than absorb.

Shift 2: Query behaviour is getting longer and more conversational

What’s happening: customers used to type “plumber Newcastle.” Now, especially on AI tools, they type or speak full sentences: “I’ve got a slow drain in my kitchen sink and the water is backing up, what’s the most likely problem and should I call a plumber or try to fix it myself?”

This is no longer hypothetical either. Google described its new Search box as the biggest upgrade in over 25 years, now accepting text, images, files, videos, and Chrome tabs as inputs simultaneously, with AI reasoning across all of them. A customer can take a photo of a broken pipe, type a question about it, and get an answer that combines both. The “keyword in a box” model isn’t being tweaked; it’s being replaced.

That’s a fundamentally different search. It contains intent, context, urgency, and a decision-making question all in one. The old SEO playbook of optimising for a single keyword phrase isn’t designed for this.

What to do about it:

Write content that answers real, specific, situational questions rather than just targeting keywords. The blog post titled “Top 5 Causes of a Slow Kitchen Drain” is still useful, but the one titled “Slow kitchen drain with water backing up – DIY fix or call a plumber?” is what AI tools are now matching against. The second version mirrors how people actually ask. It’s also where the genuine commercial intent sits.

Your service pages should do the same. Long-tail, situational, problem-framed language tends to perform better in AI retrieval than short keyword-stuffed pages.

Shift 3: Brand recall is becoming a ranking signal in itself

What’s happening: AI tools cite sources, but they cite them differently depending on how often a brand appears across the web and how authoritatively it’s positioned. A Sydney law firm that’s quoted in the Australian Financial Review, referenced on Reddit threads about Australian employment law, and appears in podcast transcripts is going to be cited more often than one that only exists on its own website, even if both rank similarly in Google.

This is where the GEO industry actually has a fair point. Brand visibility across the web, not just on your own site, is becoming a meaningful factor in AI search results.

What to do about it:

Invest in genuine brand-building outside your own website. Guest posts on Australian industry publications. Quotes in journalism (HARO is still useful; there’s an Australian version called SourceBottle worth a look). Podcast appearances. Case studies featured by suppliers or partners. Real reviews on real platforms. Industry awards.

Critically, this is earned visibility, not manufactured mentions. Google’s spam systems are smart, and so are the underlying systems most major AI tools use. Fake brand mentions on low-quality blogs will be filtered out. Real ones from credible sources won’t.

Shift 4: Search journeys are becoming conversational and multi-step

What’s happening: a customer used to do five separate Google searches over the course of an hour to research a purchase. Now they often have a single conversation with an AI tool that covers all five questions in sequence, comparing options, asking follow-ups, and refining their criteria. By the time they leave the conversation, they have a shortlist of two or three businesses to actually contact.

And it’s getting more dramatic. At I/O, Google announced Search agents – AI agents that work in the background 24/7, monitoring topics, gathering information from across the web, and sending synthesised updates back to the user. A customer planning a home renovation can now have an agent quietly track quotes, read new tradie reviews, and surface relevant content for weeks before they make a decision. Google also previewed Personal Intelligence, rolling out to nearly 200 countries, where AI Mode can draw on a user’s Gmail and Photos to personalise answers. The buying journey is no longer a series of searches; it’s an ongoing AI-mediated conversation.

The implication: you don’t get five chances to appear anymore. You get one conversation to be in. If your business doesn’t surface during that conversation, you’re not on the shortlist.

What to do about it:

Make sure your business is findable for the full set of questions a customer asks during a buying journey, not just the obvious ones. A Melbourne dental practice should rank not just for “dentist Melbourne CBD” but also for “how much does a dental check-up cost,” “do I need a referral for a dentist,” “Medicare dental coverage Australia,” and the related cluster of questions a person actually asks before booking.

Map out the real questions your customers ask in the lead-up to buying. Make sure you have content that credibly answers each one. This is where topical depth across your website starts to genuinely outperform individual high-ranking pages.

Shift 5: AI tools cite differently than Google ranks

What’s happening: Google’s algorithm ranks pages. AI tools cite sources. Those aren’t the same thing. A page can rank well but get cited rarely, or rank modestly but get cited often. The reasons aren’t fully transparent, but a few patterns are clear: AI tools tend to favour content that’s structured for citation – clear claims, direct answers, identifiable expert authorship, dates of publication, original data or examples.

The gap between ranking and citation is about to widen further. Google also announced Generative UI in Search, which is rolling out free of charge soon. Instead of returning a list of links, Search will now design custom layouts on the fly – interactive visuals, tables, graphs, simulations, even mini-app dashboards for ongoing projects like a renovation or wedding. The traditional “list of ten blue links” isn’t being modified; it’s being replaced for a growing share of queries. Pages don’t get ranked in a custom-built dashboard; sources get cited within it.

What to do about it:

Tighten how your content is structured. Lead paragraphs that state the core answer directly. Author bios that establish genuine credentials. Dated content (publish and update dates visible). Original examples, statistics, or case studies that other sources don’t have. FAQs that mirror how real questions are asked.

None of this is revolutionary; most of it is just good content hygiene. But AI tools reward it disproportionately, and much SMB content isn’t doing so.

What this actually means for your business

If I’m being direct, this is the picture:

The core work hasn’t changed. Good SEO (useful content, a technically sound website, a maintained Google Business Profile, and real authority earned over time) still drives the majority of your visibility in both traditional search and AI search. My last article wasn’t wrong. Don’t go chasing GEO hacks.

But the targets have shifted slightly. You should be measuring whether you get cited in AI answers, not just whether you rank. You should be writing for conversational, situational queries, not just keyword phrases. You should be investing in off-site brand visibility, not just on-site content. And you should structure content for citation, not just for ranking.

And the competitive bar is rising fast. Faster than most SMBs realise. Google’s own announcements at I/O confirm what those of us working in this space see every day – the surface area customers search on is being redesigned at an extraordinary pace. The SMBs who are doing the fundamentals well and paying attention to these shifts are pulling ahead. The ones who are still publishing thin commodity content and ignoring the changes are going backwards, not because they’re being penalised, but because the surface area they used to compete on is shrinking.

The honest summary

I’ll repeat what I said last time: there’s no secret AI back door. There’s no shortcut. The fundamentals matter most.

But here’s the half I didn’t fully cover before. The way customers find businesses is genuinely changing, and the bar for “good SEO” is rising as a result. Google’s I/O 2026 announcements just made that even clearer. The agencies and SMBs that succeed in the next two years will be those that continue to do the fundamentals well and layer in the new thinking that AI search demands.

That’s not GEO as a separate product. It’s SEO that has caught up with how search actually works in 2026.

If you’re trying to figure out which of these shifts apply to your business, what’s worth investing in, and what you can safely ignore, that’s exactly the kind of conversation we have with clients every week. No jargon. No pressure. Just a straight read on where you stand and what to do about it.

Frequently Asked Questions

Has AI search killed SEO?

No. The fundamentals of SEO (useful content, technical health, authority, and a strong local presence) still drive most of your visibility in AI search results. Google’s AI Overviews and Mode are built on Google’s core ranking systems. What’s changed is that the presentation layer of search now includes AI-generated answers and citations, which adds new considerations on top of traditional SEO, not in place of it.

What’s the difference between ranking and being cited by AI?

Ranking refers to where your page appears in a list of search results. Citation refers to whether AI tools reference your business or content when generating an answer. They’re related but not identical. A page can rank well and rarely be cited, or rank modestly and be cited often, depending on how well it’s structured, how credible its authorship is, and how the brand is recognised across the wider web.

Should I create content specifically for ChatGPT or Perplexity?

Not as a separate product, no. But you should be aware that ChatGPT, Perplexity, Copilot, and Google’s AI features each crawl and cite differently. Content that’s well-structured, factually accurate, expertise-led, and clearly attributed performs better across all of them. The goal is content that’s good enough to be cited by any AI tool, not content tailored to one specific platform.

How do I know if AI search is affecting my business?

Look at three signals. First, has your informational blog traffic dropped in the last 6-12 months, even though your rankings are stable? That’s often AI Overviews absorbing the clicks. Second, are you appearing in AI Overview citations for searches relevant to your business? You can check this manually by searching the queries your customers use. Third, are enquiries trending toward higher intent (fewer general questions, more specific buying-stage questions)? That usually indicates that AI is handling the earlier stages of research before customers reach you.