Before you read the numbers
- Consent and blockers. Cookie consent, ad blockers and iOS tracking limits make analytics tools undercount visits and channel attribution. Compare trends inside one tool, not across tools.
- Same word, different meaning. Active user, session and engagement differ by tool. Write down your own definition before comparing periods.
- Small counts. When counts are small, one event can swing a rate. Widen the window or watch the raw count.
- New products. With no 90-day history, use the first weeks as a baseline and reset it monthly.
- Timing. Long trials and annual plans mature slowly, so judge only cohorts whose period has fully ended. Seasonal peaks such as Black Friday distort baselines, so exclude them or compare with the same period last year.
- Mixed businesses. If you have both an app and a website, or self-serve plus sales, read each funnel on its own and check the handoff between them.
Saved in this browser only
Copying was blocked here. Select the text below and copy it yourself.
There are no industry numbers here on purpose. Benchmarks differ by price, audience and channel, so each KPI explains how to set a baseline from your own data.