You're Still Number One on Google. So Where Did the Enquiries Go?

If you rank first on Google and your enquiries have dropped anyway, there's a good chance AI search is the reason. When one of Google's AI Overviews sits above the top result, that result gets roughly half the clicks it used to, and on some local searches it gets almost none. Your ranking report won't show any of this, because your ranking hasn't changed.

I've been working on how people find local businesses online since 1996, when I was on the team that launched Citysearch in New York. In 2003 I started one of Australia's first SEO agencies, in Sydney. For most of the years since, getting to number one on Google was the whole job. Once you were there, the enquiries mostly took care of themselves.

That stopped being true faster than most owners realise. AI Overviews arrived in Australian search in October 2024, and Google switched on AI Mode here on 8 October 2025. A lot of searches now open with an answer Google has written, and the list of websites starts underneath it. When Peec AI tracked business prompts with buying intent in April 2026, an AI Overview showed up on 86.5% of the Australian ones.

Does ranking number one on Google still get clicks?

It does, but a lot fewer of them when there's an AI answer above you. Advanced Web Ranking tracks more than 1.75 million US results pages every month, and in July 2026 the top result on desktop had a click-through rate of 20.02% on a normal page. With an AI Overview above it, that fell to 9.69%. The position was identical in both cases, and the clicks dropped by 52%.

Ahrefs came at it from a different angle, comparing 300,000 keywords in December 2023 with the same keywords in December 2025, and got a similar result. Where an AI Overview appeared, the top result's click-through rate was 58% lower and the second result's was 50.8% lower.

The fair objection to both studies is that searches with AI Overviews might be different kinds of searches, so you aren't comparing like with like. Two experiments this year dealt with that properly. Saharsh Agarwal and Ananya Sen randomly split 1,065 people in the US into groups that used Google with or without AI Overviews for two weeks, which came to 68,089 searches. On the searches where an overview appeared, clicks through to websites fell by 39.8%, and the share of searches that ended without any click went up by 34.5%. A second preregistered experiment with 1,100 people, posted in August by Stephanie T. Wang and four co-authors, came to the same conclusion. When the researchers removed AI Overviews and AI Mode, more people clicked through to websites.

The part of the first study I keep coming back to is what happened to ads. Ad clicks stayed where they were, so the clicks that disappeared didn't go to a competitor's ad. People got what they needed from the answer and didn't click on anything.

Where the clicks went

They mostly end up in one of four places, and you can't see any of them from your own analytics.

The biggest share stays in the answer. Pew Research Center tracked the actual browsing of 900 US adults across 68,879 Google searches. When an AI summary was on the page, people clicked a regular result on 8% of visits, compared with 15% when there wasn't one. They clicked a link inside the summary itself on just 1% of visits, and 26% of those visits ended the browsing session then and there. Australians report much the same about themselves. In Telsyte's 2026 survey of 2,023 Australians, 81% knew AI summaries show up in search results, and half said they'd often rely on the summary without clicking through.

Plenty of searches end with no click at all. SparkToro, working with Similarweb's clickstream panel, found that 68.01% of US Google searches between January and April 2026 finished without a click on anything, and that count includes ads and Google's own properties. Some of those people would have searched again and been handed another answer.

For local businesses, the map pack is where it hurts most. According to Advanced Web Ranking, when an AI Overview and a local map pack both sit above the first organic result, that result loses 97% of its usual clicks. If you're in the map pack and named in the answer, you'll be fine. If all you've got is the top blue link, you're a long way down a search you're supposedly winning.

And a growing number of people skip Google for these questions. OpenAI reported 900 million weekly ChatGPT users in February 2026, and Google says AI Mode went past 1 billion monthly users within a year of launching. BrightLocal's February survey of US consumers found 45% had used ChatGPT or another AI tool for local business recommendations, up from 6% the year before. Those people never see your ranking.

Why doesn't my reporting show any of this?

Most SEO reports lead with rankings, and your rankings haven't moved. The rank tracker looks much the same as it did two years ago, and Search Console might even show impressions going up. What fell was clicks, and clicks don't usually get top billing in a monthly report.

People in search have started calling this the great decoupling, where impressions go up while clicks stay flat or fall. Some of it comes down to how impressions are counted. If a link to your site appears inside an AI answer, Google counts an impression whether or not anyone visits.

Google added a generative AI report to Search Console in June 2026. It's worth a look, because it tells you how often links to your site appeared in AI Overviews and AI Mode. It won't tell you about clicks, though, or which questions people asked. Those AI impressions are also already included in your normal search totals, which pushes up the number most people take as good news.

The visits you do get from AI are hard to spot as well. Someone who clicks through from an AI Overview shows up in Google Analytics as ordinary Google organic traffic. Someone who asked ChatGPT for a few names and then searched for yours shows up as a branded search, so the AI that put you on the shortlist gets no credit for it. Jayson Munday goes into the analytics side in more detail in One in Eight Australians Now Asks AI First.

If your monthly report shows steady rankings and not much else, I'd take that as a reason to look harder.

Does ranking first get me into the AI answer?

Not as reliably as it used to. In July 2025 Ahrefs found that about 76% of the pages cited in Google's AI Overviews also ranked in the top 10 for the same search. When it ran the study again in March 2026, across 863,000 searches and 4 million citations, the figure was 38%, and another 31% of cited pages didn't rank anywhere in the top 100. Ahrefs says part of that change is down to AI Overviews moving to Gemini 3 in January 2026 and part to improvements in its own parsing, so I wouldn't treat it as a precise before and after. The direction is pretty clear, though.

The overlap is smaller again outside Google. When Ahrefs looked at 15,000 prompts in August 2025, only about 12% of the pages ChatGPT, Gemini and Copilot cited ranked in Google's top 10 for the same question. We see the same pattern in our own monitoring. On the same question and the same day, ChatGPT and Google's AI Overviews had only 8% of their sources in common (the full data is here).

In practice, you can hold the top spot and still not be mentioned in the answer sitting above it.

What being named in the answer is worth

Seer Interactive has the clearest numbers I've seen on this. It tracked 2.43 billion organic impressions across 53 brands, and on informational searches with an AI Overview, brands cited in the overview got about 20,743 organic clicks per million impressions. Brands that weren't cited got about 9,445. Where there was no AI Overview at all, the figure was about 33,500.

Everyone is getting fewer clicks than before, but the business that gets named keeps more than twice as many as the one ranked next to it. By the time someone clicks through to the named business, Google has more or less recommended it to them.

What it costs you in dollars

The simplest way to price it is to ask what you'd pay to replace those enquiries. Most owners notice the phone going quiet before they notice anything in a report, and the usual response is to spend more on ads. In Australia there's decent data on what that costs.

Family law firms pay about A$210 for each enquiry they buy through Google search, based on three firms' Google Ads accounts over the 12 months to September 2026 (PixelRush). Allied health clinics pay a median of A$107 per new-patient enquiry from Google Ads (Clinic Mastery Marketing, July 2026).

To work it out for your own business, you need four numbers you probably already have:

1. How many organic enquiries you got a month a year ago, and how many you get now.
2. What a paid enquiry costs you today.
3. Your close rate.
4. What an average client is worth to you.

Multiply the drop in the first number by the second, and you've got the extra ad spend it takes to stay where you were. Multiply it by the third and fourth instead, and you've got the revenue you're missing if you haven't replaced those enquiries.

As an example, a family law firm that's lost ten organic enquiries a month would spend about A$2,100 a month, or A$25,200 a year, buying them back at A$210 each, and that only gets the firm back to where it was a year ago.

Google's side of the story

Google's position is that clicks are holding up. In August 2025 Liz Reid, who heads Google Search, wrote that "total organic click volume from Google Search to websites has been relatively stable year-over-year", and that the clicks were higher quality than before. Google didn't publish any figures to support it.

Even if that's right, it's a total across every website. If roughly the same number of clicks is being split between fewer businesses (the ones that get named in the answers), the total can stay flat while your share shrinks. It's also worth knowing that in the Agarwal and Sen experiment, people weren't any more satisfied with their searches when AI Overviews were switched on. The answers are getting the attention mainly because they sit at the top of the page.

How do I check whether AI search is costing my business?

You can get a fair idea in about ten minutes with Search Console, and an hour with the AI tools themselves will show you who's being named instead of you. This is the order I'd do it in.

1. In Search Console, go to Performance and compare the last three months with the same three months a year earlier. It only keeps 16 months of data, so a year back is as far as you can go. Filter to the pages that bring in enquiries, and look for any where average position held steady but clicks dropped.
2. Open the generative AI report and see which of your pages appear in AI Overviews and AI Mode. If one of your main money pages has no AI impressions at all, it's ranking without being cited.
3. Ask ChatGPT, Google's AI Mode and Perplexity the sort of question a customer would ask, without your business name in it, something like "who's the best family lawyer in Parramatta". Write down who gets named. Ask each one three times, because the answers change from one run to the next.
4. Look at which pages each tool cited. You'll often find reviews, directories, list articles and local news sites rather than businesses' own websites, and those are the places you need to be mentioned.
5. Add "ChatGPT or another AI assistant" as an option in the how-did-you-hear-about-us question on your enquiry form. Referral data misses a lot of AI-driven enquiries, but customers will usually tell you if you ask.

What gets a business named

Nobody can guarantee you'll be cited, but a few things make it a lot more likely.

Answer the question near the top of the page. If people ask about price, service area, how long something takes or what qualifications you hold, put that in the first couple of sentences in plain language, so it's easy for an AI tool to quote.

Give each question that brings in business its own page. "How much does a family lawyer cost in Sydney" deserves a proper answer of its own rather than a line buried in a services page.

Get mentioned in the places these tools read, which means reviews, industry directories, "best of" lists, comparison pages and local press. Almost a third of the pages cited in AI Overviews don't rank in Google's top 100, so a lot of what goes into the answer comes from pages you don't own.

Keep your details consistent everywhere. Your business name, address, services and founding year should match on your website, your Google Business Profile and every directory listing. Conflicting details give these tools less reason to be confident about you.

Then keep checking. Ask the same customer questions every month in each tool, and keep a record of who gets named, as well as whether you do.

None of this means ranking well has stopped mattering. A good share of what AI Overviews cite still comes from pages in Google's top 10, and plenty of people still click. But "we're number one" no longer tells you the whole story, and it's worth knowing what the answer says about you as well as where you rank.

Our free audit shows which businesses AI tools name when your customers ask. See who's taking your enquiries. It takes about a minute.

Sources

• Advanced Web Ranking, Google Organic CTR by Position, July 2026 data. Google US desktop, more than 1.75 million SERPs tracked a month.
• Ahrefs, Update: AI Overviews Reduce Clicks by 58%, Ryan Law, 4 February 2026. 300,000 keywords, desktop, December 2023 against December 2025.
• Saharsh Agarwal and Ananya Sen, The Impact of Google AI Overviews on Publisher Traffic and User Experience: Evidence from a Field Experiment, SSRN working paper, revised 17 June 2026. 1,065 US desktop Chrome users, 7 January to 10 February 2026. Not yet peer reviewed. Coverage in Search Engine Journal and the AEA RCT Registry entry.
• Stephanie T. Wang, Jeffrey Gleason, Yakov Bart, Christo Wilson and Danaé Metaxa, AI in Search Reduces Publisher Referrals Without Improving User Experience: Experimental Evidence, arXiv, 18 August 2026. Preregistered field experiment, 1,100 participants.
• Pew Research Center, Google users are less likely to click on links when an AI summary appears in the results, 22 July 2025. 900 US adults, 68,879 searches, March 2025.
• Telsyte, Australian Artificial Intelligence Study 2026, 2 June 2026. 2,023 Australians aged 16 and over, April and May 2026.
• SparkToro, In 2026, less than one third of Google searches still send a click, Rand Fishkin, 9 June 2026. Similarweb clickstream panel, US, January to April 2026.
• Ahrefs, Update: 38% of AI Overview Citations Pull From Top 10 Pages, Louise Linehan, 2 March 2026. 863,000 keyword SERPs, 4 million AI Overview URLs.
• Ahrefs, Only 12% of AI Cited URLs Rank in Google's Top 10 for the Original Prompt, Louise Linehan, 11 August 2025. 15,000 prompts.
• Seer Interactive, AIO Impact on Google CTR: 2026 Update, 24 April 2026. 53 brands, 2.43 billion organic impressions.
• Peec AI, analysis of 500,000 business prompts, 24 September 2026 (data from 13 to 20 April 2026).
• Google, Introducing Search Generative AI performance reports in Search Console, June 2026, and the report's help page.
• Google, AI in Search is driving more queries and higher quality clicks, Liz Reid, 6 August 2025.
• Google, I/O 2026: Welcome to the agentic Gemini era, Sundar Pichai, 19 May 2026.
• Google, Introducing AI Mode in Australia, 8 October 2025, and AI Overviews in Australia, 29 October 2024.
• TechCrunch, ChatGPT reaches 900M weekly active users, 27 February 2026.
• BrightLocal, Local Consumer Review Survey 2026, 11 February 2026. 1,002 US adults.
• PixelRush, Law Firm Cost per Lead in Australia: 2026 Benchmarks.
• Clinic Mastery Marketing, Clinic Google Ads Cost Benchmarks 2026.
• Brain Buddy AI, We Probed Five AI Engines 20,378 Times, 28 September 2026.