Every benchmark runs your prompt library through four AI engines — the ones buyers actually use to research and compare vendors. Each prompt is executed against each engine for real, and the full response is captured verbatim.

The four engines

• ChatGPT — OpenAI’s assistant, the most-used consumer AI and the default “ask an AI” for many buyers.
• Claude — Anthropic’s assistant, with live web search.
• Perplexity — an answer engine built around web search and citations.
• Gemini — Google’s assistant.

You choose which engines to include when you trigger a run; leaving all four selected gives the complete picture. Each engine is scored independently, so you can see where you’re strong on one and weak on another.

Real responses, captured verbatim

Each prompt produces a captured benchmark response from the selected engine, with returned citations and model version stored beside it. Consumer chat sessions can differ because of personalization, geography, model version, and execution path. When a report says you lost a comparison on Gemini, you can open the captured Gemini response and inspect the supporting record.

Model version on every response

AI models change often, and a model update can move your answers as much as any content you ship. So the model version is recorded on each captured response. That lets you tell apart two causes of a score change: the model changed, or the model’s answer about you changed. When you’re reading a shift in the trend, the recorded versions are how you rule out “the engine just updated.”

Engine-by-engine in your report

Because scoring is per-engine, your report breaks results down by engine as well as in aggregate. Use it to spot divergence — a competitor that only beats you on one engine, a factual error that appears on Perplexity but not ChatGPT — and to prioritize. A gap on the engine your buyers use most is worth more than the same gap on one they don’t.