Not long ago, searching for a restaurant in Sydney’s Darling Harbour meant typing something like ‘best restaurants Darling Harbour’ into Google, opening half a dozen tabs and working your way through reviews, maps, menus and recommendations.
Increasingly, that feels like the long way around.
Today, you can ask an AI tool for ‘a casual bistro near the waterfront in Darling Harbour that locals go to, open on Sundays, with good vegetarian options, nothing too touristy, and under $35 a head.’ The difference is easy to miss. The query is longer, certainly. But more importantly, the job we are asking technology to perform has changed.
We are no longer simply asking a search engine to help us find information. We are asking an answer engine to interpret what we want, weigh competing options and narrow the field for us.
That distinction has significant implications for almost every brand and business. For the past three decades, digital visibility was largely about getting found. In the next phase of the internet, getting found may only be the first hurdle. Increasingly, the bigger challenge will be getting chosen.
Search is becoming selection
The pace of this shift is striking. Similarweb data reported by TechCrunch in July 2026 found that Google’s AI Overviews were appearing in 43 per cent of searches, up from 15 per cent a year earlier. Over roughly the same period, visits to Google’s conversational AI Mode rose from 126 million in June 2025 to 279 million by May 2026.
This matters because the rise of AI search is not simply a story about people abandoning Google for ChatGPT. Google itself is becoming an answer engine. What began as an AI layer sitting above traditional search results is increasingly becoming part of the search experience itself.
Our behaviour is changing with it. Similarweb has also observed that Google queries are becoming longer and more conversational as people move away from terse keyword searches and towards natural-language requests. That makes sense. Once a system can understand context, we naturally give it more context.
Instead of searching ‘best CRM’, a buyer can ask for software suitable for a 30-person professional services firm, with strong Microsoft integrations, Australian support, a budget cap and minimal implementation complexity. Instead of searching ‘family SUV’, a customer can specify price, boot space, fuel type, safety priorities and how many child seats they need to fit across the back row.
Search engines traditionally gave us options. Answer engines increasingly help us make sense of those options.
Welcome to the shortlist economy
For businesses, this creates what I think of as the shortlist economy.
The search economy rewarded being found. The shortlist economy rewards being chosen.
That may sound like a subtle difference, but it changes the point at which influence happens. In the old model, the search engine presented a list and the customer did most of the comparing. The website, landing page, review, sales team or store experience then had the opportunity to win them over.
With AI-mediated discovery, part of that comparison can happen before a customer reaches you at all. An answer engine may summarise your offer, compare it with alternatives, draw on reviews and third-party commentary, identify likely trade-offs and decide whether you deserve a place in the response.
HubSpot’s January 2026 survey of more than 3,000 CRM purchase decision-makers offers one window into how consequential this can become. It found that 42 per cent of respondents used AI search during their evaluation process. Those buyers were 36 per cent more likely to purchase than buyers who did not use AI search. HubSpot is naturally an interested party in the growth of AEO, so its findings should be treated as company research rather than a universal benchmark. Even so, the behaviour it describes is worth paying attention to: AI is moving upstream in the buying journey, into the period when customers are still deciding which options deserve consideration.
This is why Answer Engine Optimisation, or AEO, matters. SEO has traditionally focused on helping a page rank so a human can discover and click it. AEO is concerned with something slightly different: making a brand, product or organisation sufficiently clear, relevant and credible that an AI system can understand it and potentially include it in an answer.
Put another way, SEO asks: how do we earn the visit? AEO increasingly asks: how do we earn a place on the shortlist?
Your reputation now has a machine-readable version
This is where some of the AEO conversation becomes more interesting than the acronym itself.
A customer may know your reputation from experience, word of mouth or years of seeing your brand. An AI system does not have that relationship. It has to assemble a picture from information it can retrieve or has learned: what your own website says, what customers say, what publishers and reviewers say, how consistently your business is described, whether key facts are current and whether there is enough evidence to support the claims being made.
I think of this as your machine-readable reputation.
It is the version of your organisation that can be reconstructed from the digital evidence surrounding it. And in an answer-engine world, that evidence becomes strategically important.
Consider two businesses selling similar products. One has a beautifully designed website filled with broad claims about quality, innovation and customer service. The other clearly publishes specifications, pricing guidance, availability, use cases, comparison information, frequently asked questions, independent reviews, original research and recent customer outcomes. The second business has given both people and machines far more to work with.
This is why many of the most sensible AEO practices are surprisingly unglamorous. HubSpot’s current guidance recommends making sections self-contained, clearly highlighting important facts, publishing original research, showing author and update information, and ensuring that a brand’s message is consistent across owned and third-party channels. Those ideas are not magic tricks for manipulating an algorithm. They are ways of reducing ambiguity.
And ambiguity is costly when a machine is deciding whether it has enough confidence to mention you.
The danger of chasing the algorithm
Of course, whenever a new optimisation discipline appears, an industry of shortcuts quickly follows.
One of the most revealing examples involves Reddit.
In July 2026, Yuval Halevi and the team at Growtika published the results of 8,616 API runs examining how OpenAI models handled questions about online communities. Their results were extraordinary. Across one set of 300 prompts that never mentioned Reddit, GPT-4.1, GPT-5.5 and GPT-5.6 cited Reddit in 298 answers. In Growtika’s logged searches, GPT-5.6 inserted Reddit into all 300 of its search queries for those prompts.
If you saw those results in isolation, the obvious AEO advice would be simple: get yourself onto Reddit immediately.
Then, a few weeks later, the picture changed.
Promptwatch reported that Reddit’s share of citations in its ChatGPT Search dataset had averaged 3.83 per cent between July 18 and August 7, before falling below 1 per cent on August 14. The average from August 14 to 17 was 0.52 per cent. Importantly, Promptwatch did not claim to know why. Author Klaas Foppen explicitly noted that a change in source selection was only one possible explanation and that a data-collection issue could not be ruled out.
The point is not that Reddit matters or does not matter. The point is that source preferences, retrieval behaviour and model architecture can move remarkably quickly. Different answer engines behave differently. Different model versions behave differently. And what looks like an optimisation rule in July can become a historical curiosity by August.
That makes any strategy built around gaming one platform, one source or one model inherently fragile.
A more durable approach is to optimise for the underlying decision: can an AI system understand what you offer, verify important claims and find enough credible evidence to justify including you?
The click is not dead, but it is no longer the whole game
There is another reason the old digital playbook needs updating: visibility and traffic are beginning to separate.
Chartbeat’s 2026 data shows that pageviews from Google Search to publishers fell 34 per cent between December 2024 and December 2025. The decline was particularly severe for smaller sites: search referrals fell 60 per cent for small publishers, compared with 22 per cent for large publishers.
Yet it would be premature to declare the death of the click. Similarweb data reported by TechCrunch showed that after a ChatGPT search update in May 2026, the proportion of US desktop referrals landing on webpages rose sharply, from about 25 per cent in March to nearly 60 per cent by the end of May.
That apparent contradiction is useful. AI search is not simply replacing links with answers. The relationship between answers, citations, recommendations and website visits is still being worked out in real time.
What seems clearer is that a click can no longer be the only measure of influence. A customer may encounter your brand, compare you with alternatives and form an opinion about you inside an AI conversation without ever visiting your website. Conversely, if they do eventually click through, they may arrive far better informed than a traditional search visitor.
For marketers, that means the funnel is becoming harder to see. Some of the most important persuasion may happen before your analytics platform registers a session.
The ABCs of AEO
So what does all of this mean in practice?
The good news is that succeeding in the age of AEO doesn’t require abandoning everything we know about digital marketing. High-quality content, a strong reputation and a clear value proposition matter just as much as ever. What has changed is the audience for those signals.
The question is no longer simply, “How do I rank higher?” Increasingly, it is also, “How do I become a business that AI can confidently recommend?”
I find it helpful to think about that challenge through three simple principles: Awareness, Breadth and Credibility.
A is for Awareness
Start by understanding how AI currently sees your business.
Ask ChatGPT, Gemini, Claude or Perplexity the same kinds of questions your customers might ask, without mentioning your brand by name. Which organisations appear? How is your business described? Which capabilities are recognised, overlooked or misunderstood?
Think of this as a reputation audit rather than a search audit. You are trying to understand the picture AI systems can currently assemble from the information available about your organisation.
Before you can improve your machine-readable reputation, you need to know what that reputation currently looks like.
B is for Breadth
Your website is still important, but it is only one part of your digital footprint.
Depending on the platform and query, answer engines may draw on customer reviews, media coverage, industry publications, forums, social platforms, directories and other third-party sources when constructing an answer.
That means consistency matters across the wider information ecosystem.
The goal is not to produce content everywhere. It is to ensure that credible, useful evidence about who you are and what you do exists beyond your own website.
A broad digital footprint gives an answer engine more opportunities to find, corroborate and understand your business.
C is for Credibility
Breadth makes your business easier to discover. Credibility gives an answer engine more evidence to justify recommending it.
Slogans and superlatives are not especially useful evidence. Your website can call your service exceptional, innovative or market-leading, but an answer engine has far more to work with when those claims are supported by reviews, independent coverage, awards, certifications, original research, case studies and measurable outcomes.
The useful question is simple:
If an AI system had to explain why someone should choose your business instead of another, what evidence could it point to?
The stronger that evidence becomes, the easier your organisation is to understand, verify and recommend.
Those three principles give you the strategy. The next step is turning them into habits.
Your 10-point AEO Action Plan
- Audit how AI currently sees your business. Develop a small set of realistic customer questions and test them across ChatGPT, Gemini, Claude and Perplexity. Track whether your organisation appears, which competitors appear alongside it, how you are described and which sources are being used. Tools such as HubSpot‘s free AEO Grader can provide another useful snapshot – https://www.hubspot.com/aeo-grader.
- Rewrite your homepage for clarity. Within a few seconds, a visitor should understand who you help, what problems you solve and what distinguishes your business. If someone removed your logo, would the page still make it obvious what your organisation actually does?
- Turn customer questions into useful content. Pricing. Comparisons. Risks. Timelines. Trade-offs. Expected outcomes. Every business answers the same questions repeatedly. Publish clear, honest answers to them. Increasingly, these are the same questions customers are putting to AI.
- Publish something original every quarter. Conduct a customer survey. Release benchmark data. Develop a useful framework. Share proprietary research or first-hand insights. AI can endlessly recombine information that already exists, but original data and experience give the information ecosystem something genuinely new to work with.
- Expand your digital footprint selectively. Look for credible opportunities beyond your own website: guest articles, podcasts, webinars, media coverage, industry publications, customer-review platforms and relevant directories. The objective is not omnipresence. It is independent corroboration.
- Align your online profiles. Review your website, LinkedIn presence, Google Business Profile and relevant industry listings. Are your services, positioning, locations, credentials and company descriptions consistent? Small discrepancies that humans may ignore can create unnecessary ambiguity for machines.
- Collect evidence, not just praise. Testimonials are useful, but broaden your definition of proof. Case studies, measurable outcomes, independent reviews, certifications, awards, research and third-party coverage all make your claims easier to substantiate.
- Structure content for understanding. Use clear headings, descriptive titles, concise summaries, FAQs where appropriate and logical page structures. Make important facts easy to find. Good structure helps humans scan information and helps answer engines retrieve it accurately.
- Measure recommendations, not just rankings. Rankings and traffic still matter, but they no longer tell the whole story. Re-run your customer questions periodically across several answer engines and look for patterns. Are you appearing more often? Are the descriptions becoming more accurate? Are the same competitors dominating? Are the cited sources changing? Track the direction over time rather than treating any single AI response as definitive.
- Repeat the process every quarter. AI platforms are changing quickly and so is your business. Revisit your audit, update stale content, add new evidence and correct inconsistencies. AEO is unlikely to be a one-off optimisation exercise. It is better understood as an ongoing discipline of making your business easier to understand, verify and recommend.
Being found is no longer enough
There is a tendency to treat every shift in digital marketing as another technical discipline to master. Learn the new acronym. Install the new tool. Discover the ranking factors. Optimise accordingly.
AEO will certainly develop its own tools, tactics and specialists. But I suspect the more enduring lesson is less technical.
For almost three decades, the internet trained businesses to think about visibility. How do we rank? How do we get discovered? How do we win the click? Those questions are not disappearing. But they are being joined by another: if a machine is helping the customer decide, what would give it enough confidence to choose us?
That question reaches well beyond SEO. It touches product information, reputation, customer experience, reviews, PR, thought leadership, data quality and the basic clarity of what a business stands for and delivers.
The search economy rewarded being findable. The shortlist economy will increasingly reward businesses that are understandable, verifiable and worth recommending.
Being found still matters. It is just no longer the finish line.
NOTES
Similarweb. “The 2026 Generative AI Landscape Report.” Similarweb, 2026. https://www.similarweb.com/corp/reports/2026-generative-ai-landscape/
Perez, Sarah. “Google’s AI Search Is Rapidly Becoming the Default, New Data Shows.” TechCrunch, July 27, 2026. https://techcrunch.com/2026/07/27/googles-ai-search-is-rapidly-becoming-the-default-new-data-shows/
Chartbeat. “Navigating the New Traffic Landscape.” 2026. https://lp.chartbeat.com/hubfs/Chartbeat-Report-2026Q1.pdf
Chartbeat. “Pageviews Are Down, but AI’s Impact Is Complicated.” 2026. https://chartbeat.com/resources/articles/pageviews-down-ai-impact/
HubSpot. “HubSpot AEO Data: Buyers Using AI Search Are More Likely to Purchase.” Updated June 12, 2026. https://www.hubspot.com/company-news/aeo-data-buyers-using-ai-search-more-likely-to-purchase
HubSpot. “Show Up in AI Search with Answer Engine Optimization (AEO).” 2026. https://www.hubspot.com/products/marketing/aeo-guide
Halevi, Yuval. “AI Thinks the Internet Is Reddit. We Have 8,616 Answers Proving It.” Growtika, updated July 19, 2026. https://growtika.com/blog/reddit-ai-visibility-research
Foppen, Klaas. “Reddit Citations Are Dropping in ChatGPT.” Promptwatch, August 18, 2026. https://promptwatch.com/data/reddit-citations-are-dropping-in-chatgpt
____________________________________________________
Michael McQueen is a globally recognised trend forecaster, change strategist and keynote speaker.
A bestselling author of 10 books, Michael’s latest release was named by Malcolm Gladwell and Adam Grant as one of the top five must-read new leadership books. He is a sought-after media commentator, with his insights regularly featured in Forbes, The Guardian, and CNN.
To find out more about Michael and his work, click here.
