“What sources does ChatGPT use?” The useful version of this question is not global. The sources that carry answers about sunscreen are not the sources that carry answers about ERP software. What you need is the list for your category, in your country, with an owner attached to each source so you know what to do about it. Here is how to get there, and what the list looks like.
It is a guide to building the source list for your own category, and to knowing what to do with each row.
How to find out (and the trap in doing it casually)
When ChatGPT uses web search, a response may include citations and related sources. So the naive method is: ask a question, read what is shown, done. The trap is variability. A repeated prompt may return a different answer or different cited sources. One ask is an anecdote; the source list from one ask is an anecdote too. OpenAI’s guide explains web search and citations.
The method used here: a bank of real buyer questions for the category, asked several times each across the assistants the study covers, with surfaced citations recorded and deduplicated. This gives you an observed source sample for the questions and conditions tested, not a complete list of everything that influenced an answer.
The three ownership classes
Once you have the list, every source falls into one of three classes, and the class decides what to do:
- Cannot edit. Regulators, encyclopedias, official statistics. You do not get a vote on the page. The strategy is precision and presence around it, not displacement.
- Can rewrite. Retailer product pages, marketplace listings, distributor copy, anywhere the brand or its partners control the text. These can be practical places to investigate when the brand controls the listing, subject to the channel’s approval and the evidence from the study.
- Earned. Press, reviews, forums, comparison sites, third-party guides. You influence these with substance and relationships, not with edits.
What the list looks like
From our invented sample, Solaria, a sunscreen brand sold in Mexican pharmacies (competitors Bloqsol and Dermalux):
| Source | Owner class | How to fix it |
|---|---|---|
| profeco.gob.mx | Cannot edit (regulator) | Nobody outside edits this page: be accurate, and build presence on the pages around it |
| pharmacy retailer page | Can rewrite | The brand or its distributor controls this text: rewrite the listing copy |
| dermatology reference site | Earned | Not editable directly: earned with editorial, data and relationships |
Three sources, three different actions, one decision list. That is what the list is for. In a real report, each row also carries its citation share with the denominator, so you can assess which sources appear more often in the sample. What to do with each row is a different job from reading it: the retailer page to rewrite, the editorial to commission, the mentions to earn. That is the work on the sources: /trabajo-sobre-fuentes/.
The instrument for getting that list is the same run the whole section keeps pointing at: a study with a frozen prompt bank, repetitions, and every response saved and readable. It is the first thing we run: /auditoria-visibilidad-ia/.