By Steve Merrill, Founder of WRKNG Digital | September 10, 2026
An AI assistant would rather skip your product than guess about it. Thin data reads as risk. And a model recommending to a buyer avoids risk every time.
I watched Google traffic dry up for stores that assumed their pages were fine. Same story here. The data looks complete to you and looks empty to a machine.
Here's why thin data gets you skipped, and what thin actually means.
What does thin product data look like to a model?
Thin means the field exists but the value is missing or vague. A size chart image with no text. A material listed as "premium fabric." A spec table rendered as a picture. To you it looks filled in. To a model it's blank.
The model reads the raw data, not the design. If the value isn't machine-readable, the field is empty as far as the assistant is concerned.
Why does a model skip instead of guessing?
Because guessing wrong costs it trust with the buyer. If an assistant recommends a product and the fit is wrong, the person stops trusting it. So it plays safe and names the store that gave it certainty.
Your missing field isn't neutral. It's an active reason to pick someone else.
The gaps that lose the sale most often
The usual suspects are consistent. No material or a vague one. No dimensions. No compatibility info. No structured reviews. A price that doesn't match the feed. Any one of these can drop you from the answer.
Line the data up with the Google Merchant product data specification and fill the fields with real values, not placeholders.
How to fix thin data fast
Start with your best sellers. Put every spec in a labeled field, move it into Product schema, and validate that the values are real with the Schema.org validator. Then confirm the page, schema, and feed all say the same thing.
You don't have to fix the whole catalog at once. Fix the products that drive revenue first. That's where the skipped recommendations cost you the most.
What complete data buys you
Complete, consistent data makes you the safe pick. The assistant can confirm the fit, trust the price, and recommend you without gambling.
That's the whole game now. A WRKNG audit shows you exactly which products read as thin, which fields are empty to a machine, and where your data disagrees with itself.
Further reading
Frequently Asked Questions
What counts as thin product data?
A field that exists but has a missing or vague value: a size chart as an image with no text, a material listed as premium fabric, or specs rendered as a picture. It looks filled in to you but reads as blank to a model that only sees machine-readable data.
Why do AI assistants skip products instead of guessing?
Because guessing wrong costs the assistant trust with the buyer. If it recommends a product and the fit is wrong, the person stops relying on it. So it plays safe and names the store whose data gave it certainty.
Which data gaps lose the most sales?
No material or a vague one, no dimensions, no compatibility info, no structured reviews, and a price that doesn't match the feed. Any single gap can drop you from an AI answer, so align everything to the Google Merchant product data specification.
How do I fix thin data on Shopify?
Start with best sellers. Put every spec in a labeled field, move it into Product schema, validate the values are real, and confirm the page, schema, and feed all match. Fix revenue-driving products first rather than the whole catalog at once.
Want to know if AI assistants can actually find and recommend your store? Get a free AI-visibility read on your Shopify store at WRKNG Digital. We show you exactly what ChatGPT, Perplexity, and Google AI see when they look at your products.
