Not sure where to start? Take a guided tour.
AI Site Grader
AI Query Scanner
llms.txt GeneratorAverage prompt-to-query similarity is only about 25%
Average prompt-to-query similarity was 25.24% in the query translation study.
Most of the wording changes substantially before retrieval.
The numbers behind it.
- Average similarity
- Average Jaccard similarity between prompt and search query.
- 25.24%
- Prompt-query pairs
- Captured prompt-to-query transformations.
- 11,521
The evidence.
| Measure | Result |
|---|---|
| Average similarity Average Jaccard similarity between prompt and search query. | 25.24% |
| Prompt-query pairs Captured prompt-to-query transformations. | 11,521 |
Population and method
Built from 11,521 captured prompt-to-query pairs observed in OpenAI web search calls, with 100% search-query coverage in the sampled dataset.
Where the evidence stops
This study reflects captured search-query transformations rather than every hidden retrieval step a model may perform.
Some inserted entities reflect category heuristics or market leaders rather than final recommendation outcomes.
The underlying dataset is concentrated in commercial and research prompts rather than all possible conversational traffic.
What to do with this
It explains why pages built only around the literal prompt phrasing often underperform in AI search.
Cite this fact.
Source study: Query Translation“Average prompt-to-query similarity was 25.24% in the query translation study.”
How AI Translates Your Questions - Trakkr Research · Average prompt-to-query similarity is only about 25%
How AI Translates Your Questions · Trakkr Research · Updated 22 January 2026