Skip to content
Ask AI

Ask in plain language

An answer you cannot check is only an opinion.

Each answer opens into the working behind it: the metric definition used, the exact period and segment, visit and sample counts, the query evidence cards, a confidence statement, and the limitations that apply. A figure too thin to stand on is labelled as one.

Evidence

  • Metric definition beside every figure
  • Period, segment, sample and visit counts
  • Query evidence cards per step
  • Confidence and limitations stated plainly

How it helps

What Evidence gives you

1

The definition travels with the number

‘Revenue’ arrives with the definition that produced it — which events, which currency treatment, which exclusions, read from the commercial data rather than re-derived here. Two teams reading the same answer read the same metric.

2

Sample size is stated, not implied

Visit and sample counts sit beside every figure. A segment built on a few dozen sessions is presented as exactly that, with a confidence statement to match, so a thin slice is not mistaken for a trend.

3

Limitations are part of the answer

Partial beacon coverage, an incomplete period, a bot-filtering change mid-window: the answer carries it. A caveat you have to go looking for is not really a caveat.

Connect a site today. Read tomorrow's brief instead of building it.

Connect a site and the first brief arrives with the day's changes already explained — traffic separated from bots, conversions attached to revenue, and the evidence behind every sentence one click away.

Real people separated from bots Every answer shows its evidence Reveliqo runs on Reveliqo