How each assistant behaves on your questions

Which one gives no answer, which one declines to judge, which one writes three times as much - measured, not guessed.

For: Anyone deciding which assistants to prioritise

Paste this into your assistant

Compare how the assistants behave on our questions in the latest measurement: how often each one gives no answer or declines, how long the answers are, and how many brands or names each one mentions.

What happens

1. 01Counts answers per assistant, with the share that showed no answer and the share that declined to judge.
2. 02Adds the average answer length and, where answer analysis is on, how many names each answer mentions.

Tools it calls

• getEngineStatsread only

  Per-assistant answer statistics: how many answers, how often the assistant showed no answer or refused, average answer length, entities per answer, distinct entities and verdict counts. Use it to compare how the assistants behave, not how the brand does. Defaults to the latest measurement day.

What you end up with

A per-assistant profile that explains why the numbers differ between engines.

MCP is included from Growth up, and during the 7-day trial. How to connect.