"Under $40 and carry-on size": can your product page answer that question?

Questions became sentences, and sentences have conditions

Search used to be words. "Suitcase." "Carry-on suitcase." Shoppers scanned the results and filtered them by clicking one at a time. A question to an AI is a sentence. And sentences carry conditions: budget, size, how heavy it is, even when it needs to arrive.

When Google introduced shopping in AI Mode in May 2025, it gave this example. You say "I'm looking for a cute travel bag," then narrow it to a trip to Portland in May, and AI Mode runs several searches at once to work out what makes a bag suitable for rain and long trips, then shows waterproof bags with easy-access pockets. The Shopping Graph behind it holds more than 50 billion product listings, and more than 2 billion of them are refreshed every hour.

ChatGPT works in a similar way. OpenAI's help article says that if you specify a budget of $30, ChatGPT will focus more on price. It also says product results draw on structured metadata, such as price and product description, from first-party and third-party providers. And it notes that not every relevant product will necessarily be shown.

How one question with conditions breaks apart Example query

01Price"under $40"The answer your page needsThe selling price in text, with no confusion between sale and list price
02Size"carry-on size"The answer your page needsOuter dimensions including wheels and handle, with units
03Heft"light to carry"The answer your page needsHow heavy it is when empty, as a number in kg or lb
04Arrival"I fly next week"The answer your page needsShipping cost, expected delivery, return terms

The shopper asks in one sentence, but your page has to answer each condition separately. If even one is blank, your product becomes "a product we can't confirm" for that condition.

"Carry-on approved" is a conclusion, not an answer

Here's a common mistake: printing "Carry-on OK!" in big letters on the product page and stopping there. Whether a bag can go in the cabin is ultimately decided by the shopper's airline. So the shopper, and the AI reading the page on their behalf, need numbers to check that conclusion.

Baymard Institute's research on product descriptions says the same. Dimensions matter most for physical products, especially items with strict use requirements like luggage. It recommends labeling what each measurement is and always stating units. One participant said, "'65 cm'. I don't really know how that would compare with inches." When information is missing, shoppers guess, and wrong guesses sometimes lead to unnecessary returns.

Same suitcase, two product descriptions Example lines

Copy that can't answer conditions Plenty of praise, no numbers Carry-on OK! Feather-light Roomy storage A shopper asking for "under $40, carry-on" can't confirm anything here.

Copy that answers conditions A number and unit for each condition Outer size 22 × 16 × 8 in (55 × 40 × 20 cm), wheels and handle included Empty: 5.7 lb (2.6 kg) Price $39 · free shipping Shoppers can check it against their own airline's rules.

The numbers are examples, not a real product. The right column doesn't make the call for the shopper. It gives them what they need to make it themselves.

Open a tab and give it three minutes now. Think of your best-selling product and write down three conditions a shopper might ask an AI about: "under how much," "does it fit," "can I get it by when." Then look for each number on your product page. If finding it takes more than ten seconds, shoppers and AI are likely to stall in the same place.

When the numbers live only inside images

Once you've found a number, check one more thing: is it text, or text drawn inside an image? Many stores turn size charts and shipping notes into images. People can read them fine, but text inside an image isn't treated as text by search, copy and paste, or translation.

Where the answers to each condition sit on the page

PriceUsually in text. If sale and list prices sit side by side, check that it's clear which one is actually charged.
Size and heftThe most important conditions, often trapped inside an image. Move at least one table into text.
Shipping and returnsThis answers "can I get it next week?" If it's only an image at the bottom, it's hard to find.

An answer that exists isn't the same as an answer that can be read. Keeping condition numbers in text, close to the buy button, is the safest choice.

If you want to go a step further, ask your developer about structured data. Google Search Central publishes how to describe product information in a machine-readable format. It includes properties such as price and currency, shipping details, return policy, size, material and color, and says this markup can make a page eligible for merchant listing experiences on Google Search, including the shopping knowledge panel, Google Images and popular product results. When you ask, mention that the values should match the numbers shown on the page.

One more thing. OpenAI's help article also says ChatGPT may generate simplified product titles and descriptions. If your original copy only says "feather-light," there's no number left to keep after it's rewritten.

A 3-minute check for today

Put the three conditions you wrote down next to the product page.

1 min · FindA number for each conditionFind the price, size, kg/lb or shipping number that matches each condition and note it. If there isn't one, leave it blank.
1 min · CheckText or image?Search for that number with your browser's find (Ctrl+F or Cmd+F). If it doesn't turn up, it's inside an image.
1 min · MoveJust one line in textMove the number for your most important condition, with units, into one line of text near the buy button.

Where doing this by hand gets stuck

You can do this for one product. But you have dozens, and the conditions shoppers attach differ for each one. To see how ChatGPT and Gemini answer the same question, you'd have to ask many times, changing the conditions as you go. Eventually you check a few and stop with "the rest are probably similar."

AIOudit takes that part off your hands. Paste a page URL and it looks at how shoppers and AI read that page in one pass, then lays out what to fix first.

AIOudit can't see your price competitiveness, stock, ads, seasonality or brand reputation. And which products appear for a question with conditions is ultimately up to each AI.

See how shoppers and AI read your page. Run it through AIOudit.