Most common query rewrite patterns | Trakkr Research

Research tracker page for the study How AI Translates Your Questions, focusing on the most common query rewrite patterns.

Methodology: Built from 11,521 captured prompt-to-query pairs observed in OpenAI web search calls, with 100% search-query coverage in the sampled dataset.

Summary

Benchmark data indicates that AI query rewrites are dominated by heavy replacement, list-style framing, business-term insertion, and freshness signals. The model rarely preserves the original prompt, instead transforming vague inputs into evaluative, time-bound, or geographically specific retrieval searches.

Benchmark rows

Metric Value Context
Exact match rate 0.17% Only 20 of 11,521 pairs matched exactly.
Complete rewrites 31.85% 3,670 pairs fell into the complete rewrite bucket.
Year injection rate 25.66% 2,956 queries injected a year term.
Brand insertion rate 15.24% 1,756 queries inserted brand names not in the prompt.

Ranked view

Item Value Detail
Best-of framing 20.21% The model frequently turns vague prompts into evaluative searches.
List framing 20.13% List language is one of the most common retrieval transformations.
Year injection 25.66% Freshness is a recurring rewrite habit.
Brand insertion 15.24% The model often jumps straight to likely market leaders.
Location addition 13.19% Queries are often narrowed geographically before retrieval.

Related pages

Continue through the same study cluster.

Data & Sources