Two products look the same. AI keeps naming the other one. Here's why.

Two carbon steel pans. Same price, same category, both in stock, both with decent photography. A shopper asks an AI assistant which one works on an induction stove, and it names one of them. Often the same one, over and over.

The choice isn't random, and it isn't only about who ranks first in Google. It helps to look at how the answer gets made.

An answer is built in two stages

Researchers who study this describe it in two steps: the AI first looks up web pages and product listings, then writes an answer based only on what it found.

Two stages, two different jobs for your catalog:

1. Be findable. If your product page or product feed (a file listing your products for shopping tools) isn't among what the assistant searches, nothing else matters.
2. Be easy to recommend. The AI has to write a sentence explaining why this product fits. It will pick the product it can say something specific about.

Most stores lose at stage two, and never find out, because nothing in their analytics shows an answer they weren't in.

Stage one: findable

Being findable is mostly technical housekeeping:

• The product page loads fast enough, and nothing blocks search engines and AI from visiting it.
• The product is in the places those assistants look. ChatGPT's shopping results lean on product feeds from stores and on pages it visits itself; Google's AI Overviews use Google's own saved copy of the web.
• Prices and stock are current. A product feed that says in stock when the product page says sold out gets trusted less.

None of this is exciting, and all of it is the bare minimum.

Stage two: the product that's easiest to recommend

Here's the part worth remembering. The assistant is writing a recommendation it has to support. Given two products, it favours the one whose product information lets it finish the sentence.

A listing that supports “works on induction” says something like: induction-ready, flat magnetic base, works on gas, electric and induction stoves. A listing that doesn't says: engineered for performance, chef-quality, built to last.

The second one isn't worse marketing. It just gives the AI nothing to point to.

The same goes for the qualifiers shoppers attach to questions: size (“10-inch”, “fits a small stovetop”), who it's for (“beginners”, “small kitchens”), and constraints (“under $80”, “pre-seasoned”, “no nonstick coating”). Each one either exists in your product information or it doesn't.

What breaks a tie

When two listings are equally specific, a few things tend to decide it:

• Reviews. Star ratings and review text are both a sign of trust and words an assistant can quote. Google lists aggregateRating and review among the product details it recommends.
• Product codes. gtin and mpn (the barcode number and the maker's part number) let an AI recognise that your listing and three other sites describe the same product, and combine what it knows.
• Consistency. If your website, your product feed, and the marketplaces you sell on disagree about the title, the weight, or the material, the disagreement itself is a reason to pick someone else.
• Freshness. Some AI tools trust old pages less. A listing that hasn't changed in three years competes against one updated last month.

What you can't control

Plenty. Brand familiarity, each AI's own quirks, how a given assistant weighs price, and the fact that the same question asked twice can return different products. Anyone selling you certainty about this is selling something else.

What's controllable is whether the assistant has the material to name you. That's a product information problem, and it's the one we work on.

A test you can run this week

1. Write down five questions a good customer would ask an assistant before buying from you. Include the qualifiers: budget, use case, fit.
2. Ask each one in ChatGPT and Perplexity. Record which products get named, and the reason the assistant gives.
3. For every reason given, check whether your own listing could have supported that sentence. Usually it's one missing product detail, missing across all your products.

The reasons are the useful part. They tell you exactly which detail to fix first.

If your listing could have said the same thing and a competitor still gets named, the cause may be somewhere else. We cover the other common reasons in why ChatGPT recommends your competitors, and not you.