Product analytics is the practice of tracking what users do inside your product so you can make better decisions. Not what users say they do in surveys — what they actually do. Every time a user clicks a button, completes a flow, or drops off before finishing something, that behavior leaves a trace. Product analytics is how PMs and their teams turn those traces into decisions.
The tools PMs use
Mixpanel and Amplitude are the dominant event analytics platforms — they let you track specific user actions and build funnels, retention charts, and cohort analyses. Hotjar and FullStory record session replays and heatmaps so you can watch how real users navigate your product. Google Analytics 4 is the standard for web traffic and marketing attribution. Looker and Metabase connect to your data warehouse and let analysts build dashboards that the whole company can access. Most product teams use two or three of these tools in combination.
Event tracking explained
An event is any user action you decide to track. Common examples: user_signed_up, feature_clicked, checkout_completed, onboarding_step_viewed. Good event design is the foundation of every analytics insight — if you track the wrong things, or name events inconsistently, your data becomes unreliable. PMs who understand event design can work with engineers to instrument features correctly from the start, which is far better than trying to retroactively answer a question the data was never set up to answer.
Funnel analysis
A funnel tracks how many users complete each step of a sequence. A sign-up funnel might go: landed on homepage, clicked sign up, entered email, confirmed email, completed onboarding. At each step, some users drop off. The step with the biggest drop is your biggest opportunity. Funnel analysis tells you where to focus improvement efforts rather than guessing.
Cohort retention
Cohort retention tracks groups of users by sign-up date and measures what percentage are still active in week 2, week 4, week 8, and beyond. If users who signed up in January are still active at week 8, but users who signed up in March are not, something changed — for better or worse. Improving cohort retention over time is one of the clearest signals that your product is genuinely getting better for users.
What PMs should be able to do with analytics
At a minimum, a PM should be able to define the events that need to be tracked for a new feature before it ships, build a basic funnel to measure that feature's adoption, compare cohort retention across time periods to spot trends, and explain in plain English what a metric change means and what might have caused it. You do not need to be a data scientist. You need to be fluent enough to ask the right questions and interpret what the answers mean.