How Our Pill Identifier Works — and Its Limits

An identifier is only worth using if you know what it is doing. This page explains, in plain terms, where our data comes from, how a search works, and — just as important — what this tool cannot do.

Where the data comes from

Every record on this site is built from the FDA's own published label data, distributed through DailyMed and the openFDA bulk downloads. Each drug label carries a "product characteristics" description written by the labeler: the imprint code, colour, shape, size and score of the tablet or capsule, along with the drug name, strength and dosage form. We download those labels in bulk, extract the characteristics, and organise them into a searchable database. You can read the full method, including how we deduplicate records, on our methodology page.

How a search works

When you type an imprint, the search runs in your browser against a copy of the database — your search is not sent to or stored on a server. The tool tidies up what you typed (capitals, spaces and separators do not matter), looks for an exact imprint match first, then close partial matches, and then applies your colour or shape filters if you set them. Exact matches to a complete imprint are strong; a short code such as "10" matches many medicines, and the tool will show you all of them rather than pretending there is one answer.

The limits — please read these

The safe way to use any result

Compare every characteristic — imprint (both sides), colour, shape, size and score — between the result and the pill in your hand. If all of them agree, treat the identification as a strong lead and confirm it with a pharmacist before relying on it. If anything disagrees, treat the pill as unidentified. And if someone may have swallowed an unknown pill, skip the searching and call Poison Control on 1-800-222-1222 (US) first.