Web & App Analytics

The work spans event tracking design (what to log and how to name it), tool setup (GA4, Mixpanel, Amplitude, PostHog, Segment, or some combination), funnel and retention analysis, A/B test design and readout, and dashboarding for ongoing monitoring. Done well, it shortens the feedback loop between a product change and knowing whether it worked.

This is an adult professional skill. There are no hard prerequisites, but you'll move faster with basic SQL, comfort with spreadsheets beyond pivot tables, and a feel for statistics — particularly what a confidence interval is and why a 2% lift on 500 users doesn't mean anything. Many practitioners come from marketing, product, or engineering backgrounds rather than formal data science.

The fastest path is to instrument something real. Spin up a free PostHog or GA4 account on a side project (or volunteer to clean up the tracking on a friend's app) and design the event schema from scratch. Pair that with a course on experimentation — Ronny Kohavi's writing, Lenny's Newsletter deep dives, or the Reforge analytics courses if you can access them.

Common stumbling blocks: tracking everything and analyzing nothing, confusing correlation with causation in observational data, and underpowered A/B tests that produce confident-sounding garbage. You're making progress when you can sketch a tracking plan from a product spec, look at a funnel and immediately know which step to investigate, and push back credibly when a stakeholder over-interprets a noisy result.