Best-of framing is added in one fifth of rewrites
Best-of framing was added in 20.21 percent of prompt-to-query transformations, representing 2,328 specific rewrites.
An analysis of how AI systems translate user questions into search queries revealed a consistent pattern of modifying recommendation prompts to seek definitive answers.
The numbers behind it.
- Best-of conversion
- 2,328 rewrites added best-of framing.
- 20.21%
The evidence.
| Measure | Result |
|---|---|
| Best-of conversion 2,328 rewrites added best-of framing. | 20.21% |
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
Strategists and operators must ensure recommendation content delivers a clear verdict, as AI search engines actively reform user prompts to prioritize definitive best-of results.
Cite this fact.
Source study: Query Translation“Best-of framing appeared in 20.21% of prompt-to-query transformations.”
How AI Translates Your Questions - Trakkr Research · Best-of framing is added in one fifth of rewrites
How AI Translates Your Questions · Trakkr Research · Updated 22 January 2026