Visitors, pageviews, bounce rate and visit duration

Every analytics dashboard opens with a handful of numbers. Here is what they mean in Statsdot, and how to read them.

Visitors

The number of different people who visited, counted once per day. Statsdot recognises a visitor with a code that changes every day, so the same person on Monday and Tuesday counts as two visitor-days in a weekly view. This is the price of not following anyone across days, and it applies to every cookieless tool that works this way.

Read it as "visits by distinct people per day", added up over the period.

Pageviews

Every page loaded, including several pages by the same person and reloads. Pageviews divided by visitors gives the pages per visitor: a quick sense of how much people read.

Bounce rate

The share of visitors who saw only one page that day. A high bounce rate is not always bad: someone who lands on your opening hours, reads them and leaves got what they needed. It matters on pages meant to lead somewhere, such as a landing page with a sign-up button.

Visit duration

The time between the first and the last page a visitor loaded that day, averaged. A visitor who saw one page has a duration of zero, because there is no second page to measure against. So short durations often go with high bounce rates.

Read them together, and against the previous period

A number alone says little. Statsdot compares every figure with the previous period of the same length. Look for changes, then click a row (a source, a page, a country) to see what drove them: filters apply to every number at once.

Traps

• Comparing tools. Analytics that need consent count only visitors who accept, so their numbers are lower than cookieless ones. Compare a tool with itself over time, not with another tool.
• Tiny samples. On a site with 30 visitors a week, a change of 20 % is six people.
• Averages hide. A good average duration can come from a few very long visits. Check the top pages and sources behind it.