Get named. A starter guide to AI search for ecommerce brands

Shoppers stopped searching. Now they ask.

A shopper looking for running shoes used to type "best running shoes" into Google and open six tabs. Now they ask ChatGPT, "What are the best running shoes for flat feet under $150?", and get an answer back. The answer names a few brands, explains why, and suggests where to buy.

If you're on that list, you're in the running. If you're not, the shopper never sees you. There is no page two.

The audience is already here

ChatGPT reached 900 million weekly active users in February 2026. Shoppers are among them. In NIQ's May 2026 tracker, 42% of US consumers said they had used at least one AI tool to shop in the past month.

The traffic is small, fast, and valuable

Adobe tracks more than a trillion visits to US retail sites. Here's what its data shows:

• AI-referred traffic to retail sites was up 393% year over year in Q1 2026.
• In May 2026, it was up 138% year over year. That's more than 14x its level in October 2024.
• In May 2026, those visitors converted 54% better than other traffic. They also stayed 53% longer and viewed 23% more pages.

AI is still a small slice of total retail traffic, and nobody is claiming otherwise. It is the fastest-growing slice, though, and those shoppers arrive ready to buy. They've already asked their questions, and they're showing up to check the answer.

Google is sending fewer clicks, too

This isn't only about chatbots. Between January and April 2026, 68% of Google searches ended without a click, up from about 60% two years earlier. When an AI Overview appears at the top of the results, click-through drops by nearly 60%, according to Ahrefs data cited in the same study. More and more, the answer is the destination.

Shoppers use AI to narrow the list, then they check it

AI isn't making the purchase decision. Only 11% of consumers would let AI choose for them, even in low-stakes categories. Among shoppers who used AI, 54% said they double-checked what it told them.

That's good news for ecommerce brands. AI decides who makes the shortlist, and your site decides whether you close the sale. You need to win both. The rest of this guide shows you how.

How AI decides which brands to name

AI engines don't have a secret list of favorite brands. They build each answer from what they can find and read. So the useful question is where they look.

Most of what AI says about you comes from somewhere else

In January 2026, Omniscient Digital analyzed 23,387 citations across five AI engines. About half the sources AI relied on were earned: reviews, forums, press, and other people's comparisons. Commercial sources such as retailers and marketplaces made up 30%. Only 23% came from brands' own sites.

It depends on the question, though:

• For questions about what a product does or what it works with, owned content rose to 50%.
• For customer review questions, earned sources took 82%.

The study skews toward software brands, not ecommerce. The pattern still carries over. When a shopper asks what something is, AI goes to the source. When a shopper asks whether something is good, AI goes to everyone else.

What that means for a store

Look back at our shopper's question: "What are the best running shoes for flat feet under $150?" It contains two kinds of questions.

• "For flat feet" and "under $150" are fit and spec questions. Your product pages can win these outright, if they actually answer them.
• "Best" is a reputation question. You win that one through reviews and other people talking about you.

Most stores are already working on the second one. Hardly anyone writes for the first.

AI can't use what it can't read

You can't be cited if the engine can't read your page. Adobe scored how readable US retail sites are to AI and found that product pages score lowest, at 66%, compared with 75% for homepages. So the pages that should answer shoppers' questions are the ones AI has the hardest time reading.

Part of the reason is technical. In a Vercel study of AI crawler traffic, the crawlers behind ChatGPT, Claude and Perplexity didn't run JavaScript. Anything your theme loads with scripts, such as reviews widgets, size charts and tabbed specs, may simply be invisible to them. (Gemini is the exception, because it uses Google's crawler.)

The short version

AI names brands it can read, and brands that other people vouch for. Step 1 is finding out which of those you're missing.

Step 1: Find out where you stand

You can't fix an answer you've never read. Before you change a single page, find out what AI says about your category today.

Write down the questions your shoppers actually ask

Aim for 20 to 30 questions. Don't guess them. Pull them from places where customers already use their own words:

• Support tickets and chat logs. "Does this run small?" "Will this work with my stroller?"
• Product reviews. Look for the use cases people mention: "I have flat feet and these finally work."
• Site search. What people type into your search bar is a list of questions you haven't answered. (How-to: How to find what shoppers search for on your store)
• Sales and customer service calls. The questions people ask before they buy.

Write each one the way a shopper would ask it out loud, not the way you'd write a keyword. "Best running shoes for flat feet under $150" beats "flat feet running shoes."

Ask all four engines

Put every question to ChatGPT, Claude, Gemini and Perplexity. For each answer, write down three things:

1. Were you named?
2. Who was named instead?
3. Which sources were cited? Your site, a retailer, a review site, Reddit, a publication.

Use a fresh or logged-out session so your own history doesn't color the answer. Answers change from day to day, so treat one pass as a snapshot, not a verdict.

Your absences are your roadmap

Sort the questions into three piles: questions where you're named, questions where a competitor is named and you're not, and questions where the answer is vague or wrong. The second pile is where the next three steps pay off first.

Short on time? The free Selix check runs your store across all four engines and shows you where you're missing. Free. No account. 45 seconds.

Step 2: Fix the foundations

Your product pages should be the best source on the internet for your own products. For a lot of stores, they're the hardest for AI to read.

Make sure crawlers can see your pages

Open one of your product pages, view the page source, and search for a detail that matters, like a line from your size chart or a spec. If it isn't in the source, it's probably loaded by JavaScript, and the crawlers behind ChatGPT, Claude and Perplexity may never see it. Step by step: How to check if AI can read your product pages.

The fix is to put the important content in the page itself: fit notes, materials, dimensions, compatibility, and a written summary of what reviewers say. Widgets are fine for extras. The facts a shopper needs should be in plain HTML.

Check that you aren't blocking AI by accident

Open yourstore.com/robots.txt and look for AI crawler names. Each company runs more than one bot, and they do different jobs:

• OpenAI: blocking OAI-SearchBot keeps you out of ChatGPT search answers. Blocking GPTBot only opts you out of model training, and it doesn't affect search.
• Anthropic: Claude-SearchBot and Claude-User fetch pages for search and for user questions. ClaudeBot is for training. Blocking the first two can reduce your visibility in Claude's answers.
• Perplexity: PerplexityBot.

Whether you allow training is your call. If you want to be named in answers, make sure the search bots are allowed. Check your CDN and security settings as well, since some bot-protection tools block AI crawlers without anyone editing robots.txt. Step by step: How to check your robots.txt for AI crawlers and How to check if your security settings block AI crawlers.

Clean up your product data

AI reads your catalog the way a new employee would. Give it clear facts:

• Titles that say what the product is. "Women's Stability Running Shoe, Wide Fit" tells an engine more than a model name on its own.
• Complete attributes. Size range, width, materials, weight, compatibility, care. Fill in every field your platform offers.
• Structured data. Product schema with price, availability and ratings. Many Shopify and BigCommerce themes include some by default, so check that yours is complete and accurate using a schema validator (How to check your product structured data).
• The same facts everywhere. If your site, Amazon listing and retailer pages disagree on price or specs, AI has to pick one, and it might not pick yours.

Step 3: Write the answers you're losing

This is the step most brands skip, and it's the one your product pages can win outright. Go back to the questions from step 1 where a competitor was named and you weren't. Each one needs an answer on your site.

One question, one clear answer

Put the answer where an engine will find it: near the top of the relevant page, in plain sentences. "The Stride 3 has a firm arch and a wide toe box, which makes it a strong choice for runners with flat feet." That sentence can be lifted straight into an AI answer. A bullet reading "Arch: firm" can't.

Turn your spec table into a paragraph

Spec tables are written for people who already know what they want. AI answers are written for people who don't. For each important spec, write the sentence that explains what it means for the shopper: who it's for, what it's good at, and what it's not good at. Be honest about the "not good at" part. Engines favor sources that sound like advice rather than ads.

Where to publish

• Product pages: fit, use case, compatibility, and "who this is for."
• Collection pages: a short intro that answers the category question. "How to choose running shoes for flat feet" belongs above the product grid.
• FAQs: the real questions from step 1, answered in two or three sentences each.
• Guides and comparisons: for bigger questions that don't fit on one product page.

What to skip

Don't publish pages of generic AI-written filler, stuff keywords, or write answers that dodge the question. A page that says "our shoes are perfect for everyone" gives an engine nothing to quote.

Step 4: Earn mentions outside your site

Your own pages can win the "what is it" questions. The "is it any good" questions get answered by everyone else. In the Omniscient study, earned sources supplied 82% of citations on customer review questions.

Where earned mentions come from

• Reviews that say something specific. "Great shoes" doesn't help an engine. "I have flat feet and these stopped my arch pain" does. Ask customers what they used the product for, not just for a star rating.
• Forums and communities. Reddit threads, niche forums and Facebook groups show up in AI answers often. Take part as yourself, answer questions honestly, and never post fake reviews.
• Comparison and roundup articles. Find the "best of" lists AI cites in your category and pitch the writers. Send product to reviewers who test things for real.
• Creators. Videos and posts where someone uses your product for a specific purpose give engines exactly the kind of evidence they look for.
• Marketplaces and retailers. Keep your listings complete and consistent. They're commercial sources AI reads too.

Check the sources, not just the answer

In step 1 you wrote down which sources each engine cited. Those are your target list. If the same review site or Reddit thread keeps showing up in your category, that's where to get mentioned first.

Step 5: Measure what it earned

Being named is the means. Orders are the goal.

Track answer share

Answer share is the number of your tracked questions where you're named, divided by the total number of questions you track. Count a question once, even if several engines name you. Re-run your question list on a regular schedule, because answers change as engines recrawl and competitors publish.

Follow the traffic

AI shoppers do click through. Adobe found they converted 54% better than other traffic in May 2026. In your analytics, set up a channel for AI referrals such as chatgpt.com, perplexity.ai, gemini.google.com and claude.ai, so that this traffic doesn't get lumped in with "referral" or "direct." Step by step: How to track AI traffic in Google Analytics and Shopify.

Connect answers to orders

The real question is which answers earned you sales. Tie each question you won to the page it sends shoppers to, and that page to the orders it produced. That's how you know which answers to protect, and which ones are worth writing next.

A 30-day starter plan

Week 1: Listen. Collect 20 to 30 real shopper questions. Run them across ChatGPT, Claude, Gemini and Perplexity. Record who's named and which sources are cited.

Week 2: Fix. Check robots.txt and your bot settings. Make sure your top 10 product pages show their key facts in plain HTML. Complete your product attributes and structured data.

Week 3: Write. Pick the five questions you're losing that matter most. Write a clear answer to each on the right page. Add an FAQ block to your top collection.

Week 4: Earn and measure. Ask recent customers for reviews that mention their use case. Pitch one roundup that AI already cites. Set up AI referral tracking and re-run your question list to get your baseline answer share.

Then repeat. This isn't a one-time project, just as SEO never was.

AI search for ecommerce: FAQs

Is this just SEO with a new name?

It builds on SEO but isn't the same thing. SEO gets you ranked in a list of links. AI search decides whether you're named in a single answer, and it relies more heavily on what other sites say about you. Good SEO foundations like crawlable pages and clean data still matter.

Do I need an llms.txt file?

Probably not yet. Google's John Mueller said in 2026 that "none of the AI systems use it". Your time is better spent making your real pages readable and making sure you aren't blocking AI crawlers.

How long until I see results?

It depends on the engine. Engines that search the web live can pick up a changed page once they recrawl it. What a model learned during training changes much more slowly. Re-check your questions monthly and look for a trend instead of expecting an overnight jump.

Can I pay to show up in ChatGPT's answers?

No. ChatGPT now shows ads to some users, but OpenAI says ads appear below responses, are labeled as sponsored, and don't influence the answer itself. Getting named in the answer is earned.

Does this work for Shopify and BigCommerce stores?

Yes. Everything in this guide applies to any store platform. The main platform-specific checks are your theme's structured data and whether important product details load with JavaScript.

AI names a short list. Be on it.

Five steps. Find out where you stand. Fix the foundations. Write the answers you're losing. Earn mentions elsewhere. Measure what it earned.

The shoppers are already asking. The only question is whether your brand is in the answer.

Check my store. Free. No account. 45 seconds.