SaaS Analytics Instrumentation Blueprint at Pre-Seed Stage

Analytics instrumentation at pre-seed has one job: tell you whether users are reaching your activation milestone and where they are dropping off. Everything beyond that is a distraction until you have product-market fit. Most pre-seed teams over-instrument (tracking 50 events before they understand what activation looks like) or under-instrument (tracking only pageviews with Google Analytics). This blueprint gives you the right scope — no more, no less.

📐 The Minimal Pre-Seed Event Model

At pre-seed, you need exactly 10–15 events. Each event maps to a user action that signals progress toward or away from your activation milestone.

Event	When to Fire	Why It Matters
`user_signed_up`	Account creation complete	Top of your activation funnel
`onboarding_step_completed`	Each checklist step completion (with step name property)	Identifies where users drop off
`activation_milestone_completed`	User reaches your defined activation event	Your core retention predictor
`core_feature_used`	First use of the primary product feature	Leading indicator of retention
`session_started`	User opens the app after initial signup	Return visit rate; not a pageview, a product session
`invite_sent`	User invites another user	Viral coefficient input; high-intent signal
`plan_upgraded`	User upgrades to paid plan	Conversion funnel endpoint
`subscription_cancelled`	User cancels subscription	Churn tracking; include cancellation reason as property

🏷️ Naming Conventions

Naming conventions matter more at pre-seed than they do later — inconsistent event names created early proliferate and become impossible to clean up without redeploying tracking across every surface of the product.

Standard convention: object_action (snake_case)

→ Object comes first: user_, project_, report_, subscription_
→ Action comes second: _created, _deleted, _updated, _viewed, _completed
→ Past tense for completed actions: report_exported not export_report
→ No spaces, no camelCase, no abbreviations: subscription_cancelled not SubCancelled

Add a properties object to every event with at minimum: user_id, account_id, timestamp, and one context-specific property (e.g., step_name for onboarding steps, plan_name for subscription events).

🛠️ Pre-Seed Analytics Stack

The pre-seed analytics stack has three layers:

→ Event collection: PostHog (open source, self-hostable, free tier sufficient for pre-seed) or Mixpanel (freemium, simpler setup). Both support the event model above and provide funnel analysis, retention charts, and cohort analysis out of the box.
→ Session recording: Microsoft Clarity (free, no session limit) or PostHog session recordings (included in PostHog plan). Watch 5 sessions per week for the first 3 months — recordings tell you what your data does not.
→ Database queries: For specific metrics your analytics tool does not expose, query your database directly. Activation rate, time-to-activation, and cohort retention are often faster to calculate from your own database than from a third-party tool at pre-seed volumes.

📊 The Three Metrics That Matter at Pre-Seed

With your event model in place, calculate only these three metrics until you reach product-market fit:

→ Signup-to-activation rate: (users who fired activation_milestone_completed within 7 days) / (total signups). Target: 30%+ for self-serve B2B. Below 20% means the onboarding flow needs work before any other growth investment.
→ D30 retention: (users who had a session in week 4) / (users who activated in week 1). Target: 40%+ for a product with weekly-use patterns. This is your product-market fit signal — not activation rate.
→ Time to activation: Median time from user_signed_up to activation_milestone_completed. Target: under 24 hours. Users who activate after 48 hours churn at significantly higher rates.

What to Do Next

If you have no analytics: install PostHog today and implement the 8 events in the table above. Do not add more events until you have data from 50+ users — the data will tell you what else you need to track. If you have analytics but many events: identify your activation milestone event and build a funnel from signup to that event. The funnel will show you your drop-off points; fix the biggest drop-off point before adding new tracking. The most common pre-seed analytics mistake is adding more events instead of acting on the events you already have.