How is AI changing the way people discover and buy products?
For most of the last two decades, ecommerce discovery followed a predictable path. A shopper typed a query into Google, scanned a page of blue links and shopping cards, clicked a few, compared options across tabs, and eventually landed on a product page. Brands optimized for that path. They fought for rankings, bid on ads, and tuned product titles to win the click.
That path is now splitting in two. A growing share of shoppers start their research inside an answer engine instead of a results page. They ask Google's AI Overviews, Perplexity, ChatGPT, or Gemini questions like "what is the best budget standing desk" or "which protein powder is good for sensitive stomachs," and they expect a synthesized answer with specific products named, not ten links to sift through. The AI reads across many sources, decides which products and brands are credible, and presents a short list. For the shopper, the comparison work is already done. For the brand, the entire game has changed, because being on page one no longer guarantees you are in the answer.
This is the shift ScaleCart, an ecommerce platform, came to RankJoe to navigate. They could see their categories generating AI answers, and they could see those answers naming other stores.
The real problem: invisible inside the AI answer
ScaleCart was not failing at traditional SEO. Their product pages ranked, their category pages pulled traffic, and their paid campaigns converted. The problem was newer and quieter. When a high-intent shopper asked an AI engine for a product recommendation in one of ScaleCart's core categories, the answer would confidently name competitors, link out to review sites and buying guides, and skip ScaleCart entirely.
That is a dangerous kind of invisibility. The shopper asking an AI "which one should I buy" is at the bottom of the funnel, close to a purchase decision, and highly receptive to a recommendation. If the AI never mentions your store, you do not get a worse ranking, you get no mention at all. There is no second page to be found on. The brand simply does not exist in that conversation.
Worse, the reasons for being left out were not obvious from inside ScaleCart's own analytics. Their pages looked fine to a human. But AI systems do not read a product page the way a shopper does. They look for structured, machine-readable signals about what a product is, what it costs, whether it is in stock, how it is rated, and whether the brand behind it is a recognized entity. Where those signals were thin or missing, the AI had nothing solid to cite, so it reached for sources that gave it cleaner, more confident data.
What RankJoe actually did for ScaleCart
RankJoe treated this as a generative engine optimization problem first and a traditional SEO problem second, then made the two reinforce each other. The work started with structure. RankJoe implemented and corrected product schema and review schema across ScaleCart's catalog so that price, availability, ratings, and product attributes were expressed in a format AI systems can parse without guessing. Clean, accurate structured data gives an answer engine something it can quote with confidence, which is exactly what it needs before it will cite a source.
Next came content built for the questions shoppers actually ask. Rather than thin product descriptions, RankJoe developed buying-guide and comparison content that answered real purchase questions directly and early, in the answer-first style that AI engines prefer to lift from. A guide that opens by clearly stating which product fits which need is far more citable than one that buries the answer under three hundred words of throat-clearing.
The third pillar was entity signals. RankJoe worked to establish ScaleCart and its key products as recognized entities, consistent across the site, structured data, and external references, so that AI systems could connect the brand to its categories with confidence rather than treating it as an unknown. Together these moves did not chase a keyword. They gave the machines a reason to trust ScaleCart as a source worth naming.
Why this gets products cited in AI Overviews and Perplexity
Answer engines are not ranking pages, they are assembling answers. To do that, they pull facts from sources they can read cleanly and trust, then attribute the result. A brand gets cited when it makes itself the easiest, most reliable source for the specific claim the AI is trying to make. If a shopper asks for the best option under a certain price, the engine needs a source that clearly states the product, the price, and why it fits. If that source is your structured data and your answer-first guide, you get named.
This is why the combination RankJoe built matters. Product and review schema give the engine hard facts it can quote. Buying-guide content gives it the reasoning and context to frame a recommendation. Entity signals tell it the brand is real and relevant to the category. Each piece lowers the friction of citing ScaleCart and raises the confidence the engine has when it does.
Perplexity and Google's AI Overviews behave slightly differently, but both reward the same underlying qualities: clarity, structure, freshness, and trust. By optimizing for those qualities rather than for any single engine, RankJoe positioned ScaleCart to show up as AI shopping behavior continues to spread across platforms.
How SEO and GEO worked together, not against each other
A common misconception is that generative engine optimization replaces SEO. In ScaleCart's case it did the opposite. The structured data, the entity work, and the answer-first content all strengthened traditional search performance at the same time they improved AI visibility, because clean structure and genuinely useful content are what classic ranking systems reward too.
The relationship runs both ways. Strong organic presence and a credible link and citation footprint are signals AI engines lean on when deciding which sources to trust. A page that already ranks well and is referenced across the web is a safer thing for an AI to quote. So the SEO foundation made ScaleCart a more attractive source for AI answers, and the GEO work made the same pages clearer and more authoritative for traditional search. RankJoe ran them as one program rather than two competing budgets.
The practical result was that ScaleCart did not have to choose between optimizing for the search of today and the search of tomorrow. The same investments served both.
What changed for ScaleCart
The most meaningful shift was that AI answers stopped being a blind spot and started becoming a channel. Where AI engines had previously named other stores in ScaleCart's core categories, the brand and its products began surfacing inside those answers for high-intent product searches. That turned the AI response itself into a discovery surface that fed real ecommerce demand rather than routing it to competitors.
Just as important was the quality of attention this attracted. Shoppers who arrive after an AI has effectively pre-qualified them and recommended a product tend to arrive with intent already formed. They are not at the top of the funnel browsing options, they are closer to the decision. Showing up in that moment connects the brand to demand at the point where it is most likely to convert.
Internally, the change also gave ScaleCart a clearer view of an emerging part of their market. Instead of treating AI search as an unknowable black box, they now had a structured, intentional presence in it, and a framework for keeping that presence current as the engines evolve.
Lessons for other ecommerce brands
The first lesson is that ranking is no longer the same as being found. You can hold strong positions in traditional search and still be completely absent from the AI answers your highest-intent shoppers are reading. Auditing your visibility inside AI Overviews and Perplexity is now a distinct exercise, separate from checking your rankings.
The second lesson is that structured data is foundational, not optional. Accurate product and review schema is the raw material AI engines quote from. If your catalog does not express price, availability, ratings, and attributes in a clean, machine-readable way, you are asking the AI to trust a competitor's data instead of yours.
The third lesson is that content has to answer the question first. Buying guides and comparisons that state the recommendation up front, backed by real detail, are far more likely to be lifted into an answer than pages that bury the point. And the fourth lesson is to treat SEO and GEO as one program. The brands that win in AI shopping are not the ones who abandon search fundamentals, they are the ones who extend those fundamentals into a format the machines can read, trust, and cite.