Setting Up Analytics That Actually Help You Decide

You have Google Analytics installed. You check your daily visitors. You feel good when the number goes up and bad when it goes down. None of this helps you make better product decisions.

The Metrics That Matter

**Activation rate.** What percentage of signups complete the core action that delivers value? If you are a project management tool, activation means creating a project and adding a task. If you are an analytics product, activation means connecting a data source.

Track this daily. It tells you whether new users are finding value. If activation is below 30%, your onboarding is broken.

**Retention.** What percentage of users who activated are still using the product after 7 days? 30 days? 90 days? This is the single most important metric for SaaS. No amount of acquisition fixes bad retention.

**Revenue per user.** Total monthly revenue divided by active users. This tells you whether your pricing is working. If ARPU is flat while you add features, you are not capturing the value you are creating.

The Anti-Metrics

**Total signups.** Meaningless without activation data. 10,000 signups and 100 active users is worse than 500 signups and 200 active users.

**Page views.** Tells you almost nothing about product health. A blog post going viral gives you a spike that has nothing to do with your product.

**Time on site.** More time can mean engagement, or it can mean confusion. Without context, this metric is useless.

The Tools

**Plausible or Fathom** for website analytics. Privacy-friendly, lightweight, and they show you what you need without the complexity of Google Analytics.

**PostHog or Mixpanel** for product analytics. These let you define events, build funnels, and track user paths through your application.

**A spreadsheet** for business metrics. Revenue, churn rate, ARPU, CAC — track these weekly in a simple spreadsheet. You do not need a dashboard for 100 users.

The Weekly Review

Every Monday, review three numbers: new activations, 7-day retention of last week's cohort, and revenue. If all three are stable or growing, keep executing. If any one is declining, investigate before doing anything else.

This takes 15 minutes. It replaces hours of dashboard-staring with actionable insight.