Compare web analytics tools

How these comparisons are built

Both tools in a comparison go through the same rows. The values come from one snapshot of the directory, and each carries the date of its own check: script sizes we measure, prices read from the vendor’s pricing page, feature values from the vendor’s documentation, plans, policies and repository. No row is weighted, and nothing adds up to a total.

The count on each tile compares the recorded values one by one. A row counts as different when the two values are not identical, and “Not confirmed” is a value like any other, so part of a difference can be a gap in our records rather than a gap in a product. “Partly” marks a feature that has limits or arrives only on some plans, and the tool card names which.

Prices compare only inside one unit. A vendor billing pageviews and a vendor billing events can publish the same figure for very different traffic, so each price shows its unit and nothing is converted. The methodology sets out every column.

Questions and answers

Why does no comparison name a winner?

Because the rows answer different questions, and only you know which of them binds. A cheaper plan is the wrong answer when data has to stay in one place, and a lighter script matters less than an import you cannot do without. The page on why there is no overall score sets out the reasoning.

How do I compare two tools that have no page here?

Open both tool cards: they hold the same rows as a comparison, in the same order. For one tool against its whole category, use its alternatives page, such as alternatives to Matomo.

How current are the values?

Each value carries the date it was checked, and they are not all checked on the same day, so one comparison can mix dates. The methodology explains how to report a value that has changed.