Start simple, expand later
When you first set up Analytiq, don’t try to track everything at once. Start with 5-7 core events that directly answer your most important questions, then expand gradually. The 5 events every app needs:| Event | When to fire |
|---|---|
page_view | Every page load / route change |
signup_completed | New user registers successfully |
login | User logs into existing account |
purchase | Payment confirmed |
logout | User signs out |
How to decide what to track
Ask yourself: “What decision would I make if I knew this number?” Examples:- “If I knew which CTA button converts better, I’d run a proper A/B test” - track both buttons separately
- “If I knew what % of signups upgrade within 7 days, I’d know if onboarding works” - track
signup_completedandupgrade_clickedwith timestamps - “If I knew which feature is most used, I’d focus work there” - track feature interactions
Full starter event list by category
Authentication
Navigation
Revenue
Feature usage
Key actions
E-commerce (if applicable)
Naming conventions
Consistent naming makes your dashboard readable. Use this pattern everywhere:| Good | Bad |
|---|---|
signup_completed | signup |
page_viewed | PageView |
button_clicked | click |
purchase_completed | payment_success |
feature_used | used feature |
plan_upgraded | upgrade |
- All lowercase
- Use underscores (
_), not spaces or hyphens - Verb at the end in past tense:
_clicked,_viewed,_completed,_started,_failed - Be specific:
signup_completedis better thansignup
Properties — what to include
Add properties to give context to each event. Properties let you filter and compare on the dashboard.| Type | Example |
|---|---|
string | 'pro', 'email', '/dashboard' |
number | 49, 3, 1024 |
boolean | true, false |
null | null (for “not set”) |
What NOT to track
| Don’t track | Why |
|---|---|
| Passwords, credit card numbers | Security risk — never in event properties |
| Full email addresses | Privacy concerns, GDPR |
| Personally identifiable info (name, address, phone) | Use identify(user) instead |
| Every single keypress | Too much noise, meaningless data |
| Mouse movements | Zero business value |
| Events with identical properties every time | Adds noise, not signal |
How many events is too many?
There’s no hard limit, but as a rule of thumb:- 5-10 events - clean, focused, easy to read
- 10-25 events - good coverage of your app
- 25-50 events - advanced tracking, useful for large apps
- 50+ events - only if each event has a clear business purpose
Sample event plan for a SaaS app
| Stage | Events |
|---|---|
| Awareness | landing_page_viewed, pricing_page_viewed |
| Acquisition | signup_started, signup_completed, google_oauth_clicked |
| Activation | first_project_created, first_event_tracked, dashboard_viewed |
| Retention | login, feature_used, report_generated |
| Revenue | upgrade_clicked, checkout_started, purchase |
| Referral | invite_sent, share_clicked |
| Churn risk | subscription_cancelled, downgrade_requested |