For years, marketers obsessed over one very familiar question.

How do we get customers to find us?

The answer used to be reasonably clear. Rank on Google. Run paid search. Build content. Collect reviews. Optimise product pages. Win attention on social. If someone had a problem, they typed a few words into a search bar, scanned the results, clicked a link, and eventually landed on your website, where your carefully written landing page waited like a polite salesperson wearing sensible shoes.

That world is not disappearing overnight, but it is changing fast.

The new question is not just, “Can customers find us?”

It is, “Will the machine recommend us?”

That shift may sound small, but for marketers it is enormous. Because increasingly, customers are not simply searching. They are asking. They are comparing. They are prompting. They are letting AI Overviews, AI Mode, ChatGPT, Gemini, Perplexity and shopping assistants compress research, filter choices and make decisions faster.

In the old search journey, brands had to win the click.

In the new AI search journey, brands may first need to win the answer.

And that changes almost everything.

From search results to suggested answers

Traditional search gave customers a list of options. AI search gives them a synthesis.

That difference matters psychologically. A list feels open-ended. It invites browsing. An answer feels more authoritative. It creates a sense of completion. Even when users know AI can be imperfect, the experience still reduces cognitive effort.

And cognitive effort is one of the quiet forces behind consumer behaviour. People like shortcuts. They like fewer decisions. They like the feeling that someone, or something, has done the hard work for them.

If you search for “best running shoes for flat feet” on a traditional search engine, you might get ads, reviews, product pages, comparison articles and some energetic SEO content written by someone who has clearly never run anywhere except toward a deadline.

If you ask an AI assistant the same question, it may summarise the key factors, compare options, explain trade-offs and narrow the decision before you ever visit a brand website.

That is both convenient and terrifying, depending on whether your brand appears in the answer.

The new middleman is the recommendation layer

Marketers have always dealt with intermediaries. Retailers, marketplaces, media platforms, comparison sites, influencers and search engines have all shaped customer choice.

AI search introduces a new kind of intermediary: the recommendation layer.

This layer does not simply display information. It interprets it. It pulls from product data, reviews, structured content, credible sources, brand pages, publisher articles, merchant feeds and shopping platforms. Then it decides what is useful enough to present to the user.

That creates a new marketing challenge. Your brand may have excellent products, but are they clear enough for AI systems to understand? Your website may look beautiful, but does it explain what you sell, who it is for, why it is different and what proof supports those claims? Your product pages may be persuasive to humans, but are they structured enough for machines?

In other words, AI search rewards brands that are not just emotionally persuasive, but machine-readable.

This does not mean marketers should start writing like robots. Please don’t. The internet has already suffered enough. It means the brand’s story, product details, proof points and trust signals need to be consistent, structured and easy to interpret.

If your website says one thing, your product feed says another, reviews highlight something different and your social content is mostly vibes with a discount code, AI systems may struggle to understand what your brand should be recommended for.

That is a problem.

The real-world shift is already happening

This is not a distant future discussion. It is already happening in retail.

Target is a useful example because it shows how quickly AI-powered discovery is moving from theory into behaviour. Target has said AI-driven traffic to its site jumped sharply in Q1 compared with the previous year, and the company is building shopping experiences across conversational environments including Google Search with AI Mode, Gemini, Microsoft Copilot and ChatGPT.

In simple language, Target is not waiting for customers to arrive through the old website journey. It is trying to appear wherever shoppers are already asking for ideas, comparisons and recommendations.

That matters because the customer journey is becoming less linear. A shopper may not start with “Target.com” anymore. They may start by asking an AI assistant, “What should I buy for a baby shower under $75?” or “What is a good dorm room setup for a university student?” If Target appears naturally in that answer, it has entered the decision before the customer has even thought about opening a retailer’s website.

Shopify is another important example, especially for smaller and mid-sized brands. Shopify says product data from merchants, including pricing, inventory, images and variants, can be surfaced through ChatGPT product discovery. That means AI shopping is not only a game for giant retailers. It may also favour smaller brands whose product data is clean, current and easy to understand.

That is a huge shift. A beautifully designed online store still matters, but if the product information behind it is vague, outdated or inconsistent, the brand may struggle to appear when AI systems are narrowing choices.

Walmart points to where this may be heading. OpenAI has described a Walmart experience inside ChatGPT that connects discovery with account linking, loyalty and Walmart payments. That is not just search. That is a compressed shopping journey. The customer is having a conversation, refining intent, comparing options and moving closer to purchase within the same environment.

Google’s Shopping Graph tells the same story from another angle. Product information sent through Merchant Center and Manufacturer Center can help power experiences across Google surfaces, including generative AI features such as product recommendations, review summaries and buying guidance.

The old marketing question was, “Can customers find us?”

The new question is, “Can AI systems understand us well enough to recommend us?”

That is the uncomfortable but important shift.

Why trust signals matter more than ever

AI search changes the role of trust.

In traditional marketing, trust was built through repetition, brand recognition, reviews and customer experience. Those still matter. But in AI-mediated discovery, trust also has to be visible in the information ecosystem.

Reviews matter. Expert mentions matter. Clear product specifications matter. Transparent pricing matters. Updated availability matters. Consistent claims matter.

A brand cannot simply call itself “premium,” “sustainable,” “customer-first” or “best-in-class” and expect the machine to nod politely. AI systems need evidence. Customers do too, of course, but customers can sometimes be emotionally persuaded. Machines are less impressed by your brand manifesto unless it connects to observable signals.

This is where marketing becomes more disciplined.

If a skincare brand claims to be dermatologist-recommended, where is the proof? If a travel brand claims to be family-friendly, do the reviews support that? If a software company claims to save teams time, are there case studies, product pages and credible comparisons that reinforce the same message?

AI search makes vague positioning weaker and specific proof stronger.

That is probably a good thing, even if it makes life harder for anyone still using the phrase “unlock your potential” as a strategy.

SEO is not dead, it is growing another head

Every few years someone declares SEO dead, usually right before selling a course about the next version of SEO.

SEO is not dead. It is mutating.

For years, marketers optimised for keywords, rankings and clicks. Now they also need to think about answer engine optimisation, generative engine optimisation and AI visibility. The terminology is still evolving, because naturally the marketing industry has responded to uncertainty by creating more acronyms. We cope how we can.

The idea itself is simple. Brands need content and data that AI systems can confidently understand, summarise and recommend.

That means answering real customer questions clearly. It means publishing useful comparisons, buying guides, FAQs, specifications, category explainers and evidence-backed claims. It means making sure product feeds, schema, reviews, website copy and customer language all tell the same story.

This is especially important in categories where customers compare before buying: travel, insurance, software, electronics, beauty, health products, education, finance, home improvement and retail.

If AI systems become the first filter, your content has to help them filter correctly.

Brand positioning now needs to be machine-readable

This may be the most important strategic shift.

In the past, brand positioning lived mainly in creative work, campaigns, websites and internal decks. Now positioning also needs to be reflected in structured, findable, machine-readable signals.

That does not make brand less important. It makes brand more operational.

A clear positioning statement is no longer enough. The brand has to be expressed consistently across product data, reviews, help pages, comparison content, social proof, partnerships and customer language.

Think about a brand like Patagonia. Its positioning around environmental responsibility is not just a slogan. It appears through product choices, repair programs, activism, storytelling and public commitments. That makes the brand easier for humans to understand, and easier for machines to interpret.

Or consider Costco. It does not need poetic brand language to communicate value and trust. Its model, membership structure, pricing perception and customer reputation all reinforce the same idea.

AI search is likely to reward that kind of coherence.

Not because machines love good branding, but because coherent brands produce clearer evidence.

What marketers should actually do

The practical lesson is not to abandon creativity and become a spreadsheet with a logo. The lesson is to connect creativity with clarity.

Marketers now need to ask sharper questions.

What do we want to be recommended for? What proof supports that recommendation? Where does that proof appear? Is our product information accurate and structured? Do customer reviews reinforce our positioning? Are we answering the questions customers actually ask? Would an AI assistant understand why we are different?

These questions are not glamorous. They will not make anyone’s agency reel look dramatic. But they may determine whether the brand is visible in the next version of search.

This is where content strategy becomes more important, not less. Brands need explainers, comparisons, use-case pages, FAQs, customer stories and expert-backed content. Not as filler. As evidence.

The future of search may reward brands that are genuinely useful.

How inconvenient.

The danger of being invisible in the answer

The scary part of AI search is not that customers will stop discovering brands. It is that they may discover fewer of them.

When AI compresses options, it compresses opportunity. If an assistant gives three recommendations, the fourth brand may as well be standing outside the shop waving through the window.

This creates a new visibility problem. Brands that once survived by ranking somewhere on page one may find that AI summaries reward only the clearest, most trusted or most frequently referenced options.

That does not mean smaller brands are doomed. In fact, niche brands with strong positioning and clear evidence may benefit. A smaller brand that clearly owns a specific use case may be easier to recommend than a larger brand with vague messaging.

But brands that rely on general awareness without clear proof may struggle.

AI search does not eliminate branding. It punishes fuzzy branding.

The human still matters

It would be a mistake to assume AI search removes emotion from marketing.

Humans still make the final decision. They still respond to story, identity, trust, aesthetics, humour and social proof. But the path to that decision is changing. The customer may arrive better informed, more filtered and already influenced by what the AI system chose to include or exclude.

So marketers need to win twice.

They need to be understood by the machine and desired by the human.

The machine needs clarity. The human needs meaning.

The brands that manage both will have a serious advantage.

If you enjoyed this psychology-led view of how customer decisions are changing, you may also like this article on how brands use fear and anticipation in marketing:

https://loyaltyandcustomers.com/articles/how-brands-use-fomo-to-win-customers-real-examples/

Final thought

AI search is not just a technology change. It is a behavioural change.

Customers are learning to outsource parts of discovery, comparison and decision-making to intelligent systems. That means brands need to think beyond rankings and clicks. They need to become credible enough, clear enough and useful enough to be recommended.

The future of brand discovery may not belong to the loudest brand or the one with the biggest media budget.

It may belong to the brand that machines can understand and customers can trust.

And if that sounds less exciting than a viral campaign, it probably is.

But it may be much more important.