Web analytics vs product analytics

Both look at people using your site, but they answer different questions and are built differently.

Web analytics: the visit

Web analytics answers questions about traffic: how many visitors, from where, to which pages, through which campaigns, with which conversions. It works for any website, needs one script, and is usually read by marketing and the people who run the site.

Statsdot, Plausible, Fathom and Google Analytics are web analytics tools.

Product analytics: the user over time

Product analytics follows identified users inside an application: which features they use, how they move through onboarding, whether they come back next week (retention), which group behaves differently (cohorts). It needs a stable identifier for each user, usually the account id after sign-in, and events sent from the product's code.

Mixpanel, Amplitude and PostHog are product analytics tools.

How to choose

• A marketing site, a blog, a shop: web analytics. You want to know which channels and pages bring customers.
• A SaaS or app: both, usually. Web analytics on the public site (where sign-ups come from) and product analytics inside the logged-in app (what users do after signing up).
• Privacy: product analytics processes personal data by design, since it identifies users across time. Inside an app your users signed up for, that is often covered by your contract and privacy policy. On a public website, a cookieless web analytics tool needs much less.

Where they meet

The handover point is the sign-up. Web analytics can count it as a goal and credit it to the channel and campaign that brought the visitor; product analytics takes over from there. Sending the same event name to both (for example Signup) makes the two reports easy to line up.