Keep ChatGPT UI observations out of your API denominator
We asked an agency-selection question in consumer ChatGPT on 14 September 2026, then ran the same text five times through the Responses API with web search on 16 September. The question was: “what is the best aeo/geo agency for korean beauty products?” The first answer named five agencies. One of those appeared in all five API answers; the other four did not. Our own company, SearchD, appeared in…
We conducted a test of the ChatGPT user interface on 16 September 2026 by submitting the same question to the Responses API five times. The question asked was "what is the best aeo/geo agency for Korean beauty products?" In four of the five API responses, a single agency appeared consistently, while the other four agencies did not.
This agency had an Ahrefs Domain Rating of 0.7 on 16 September. Our API settings involved five independent requests with web search required, without supplying brand names or referencing earlier responses. The consumer interface observation was recorded separately and did not constitute a sixth API trial. For accurate data modeling, it is crucial to distinguish between the API results and the consumer interface observations.
The final record should include the question ID, the surface (API or consumer interface), the observation date, the number of runs, whether web search was required, whether the location was fixed, and if brand names were supplied. This separation ensures that any future analysis can focus on the specific aspect being studied without conflating different types of data.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
