MVP Analytics Setup: The 12 Events to Track in the First 30 Days After Launch (and the Dashboard Founders Actually Need)

What "MVP analytics" means

MVP analytics is the small set of events and charts that tell you whether people sign up, get value from your app, come back and pay. It is not a data warehouse. It is a few answers you can check every Monday. An event is one thing a user did at one moment, with a name, for example signup_completed. A property is extra detail attached to that event, like platform: ios or source: linkedin. The usual mistake is to track everything. Most tools can log every tap automatically, so you end up with hundreds of events and no answers. Twelve named events chosen before launch work better.

The 12 events to track

Name each event once, in object_action format, and use the same names on web, iOS and Android. One chart then covers all three platforms.

• 1. first_open (mobile) or landing_viewed (web). Stage: acquisition. The question it answers: how many people arrive, and from where?
• 2. signup_started. Stage: acquisition. The question it answers: how many try to create an account?
• 3. signup_completed. Stage: acquisition. The question it answers: how many finish?
• 4. onboarding_step_completed. Stage: activation. The question it answers: which onboarding step loses people?
• 5. onboarding_completed. Stage: activation. The question it answers: how many get through the whole setup?
• 6. core_action_completed. Stage: activation. The question it answers: how many reach the moment your app is built for?
• 7. session_started. Stage: engagement. The question it answers: are users coming back?
• 8. feature_used. Stage: engagement. The question it answers: which features get used, and which get ignored?
• 9. invite_sent. Stage: engagement. The question it answers: do users bring other users?
• 10. paywall_viewed. Stage: revenue. The question it answers: how many see your offer?
• 11. checkout_started. Stage: revenue. The question it answers: how many try to pay?
• 12. purchase_completed. Stage: revenue. The question it answers: how many pay, and how much?

Acquisition: events 1 to 3

Attach a source property to first_open and signup_completed. On web, read it from the UTM tags in the link. On mobile, if you have no attribution tool yet, ask "How did you hear about us?" during onboarding. That one question at signup often gives a v1 its cleanest source data. The gap between signup_started and signup_completed tells you whether the signup form is the problem. If lots of people start and few finish, check the form before you touch your ads.

Activation: events 4 to 6

Add a step_name property to onboarding_step_completed. One funnel chart then shows which screen loses users. Event 6 matters most. core_action_completed fires the first time a user gets the result your app promises. Choose it before launch and write it down:

• Booking app: first booking confirmed.
• Habit tracker: first habit logged.
• Invoicing tool: first invoice sent.
• Marketplace (buyer side): first order placed.

Engagement: events 7 to 9

session_started gives you weekly active users and retention. feature_used takes a feature_name property, so one event tracks ten features. After 30 days, the features nobody opens are the first ones to cut or rethink in v1.1. invite_sent only matters if your app has sharing or team invites. If it doesn't, swap it for the action that best shows intent in your app, such as item_saved or notification_enabled.

Revenue: events 10 to 12

Track all three, even with only a few paying users. The step from paywall_viewed to purchase_completed tells you whether your pricing or your paywall needs work. Add plan and amount properties to purchase_completed. Stripe and RevenueCat already record payments. Send purchase_completed to your analytics tool anyway, so revenue sits in the same funnel as signups and activation.

User properties to set once

Events say what happened. User properties say who did it. These five properties cover most of what you'll want to know in the first month. Keep personal data you don't need, like full names, out of your analytics tool. Set these at signup and update them when they change:

• platform (web, ios, android)
• signup_source
• plan (free, trial, paid)
• signup_date (used for cohorts)
• country

Not sure which action is your core action?

Not sure which action is your core_action_completed? Book a call and talk it through.

The dashboard founders actually need

One page, five charts. If a chart doesn't change a decision, remove it. Use a single tool for all five. PostHog, Mixpanel, Amplitude and Firebase Analytics can all build these charts. Pick the one your developer already knows that works on every platform you ship. Crash tracking is a separate job, covered in our article on Sentry vs Firebase Crashlytics for iOS. The five charts:

• Signup funnel. Built from events 1, 2 and 3. What you decide with it: fix the store page, landing page or signup form.
• Activation funnel. Built from events 3, 4, 5 and 6. What you decide with it: fix or shorten onboarding.
• Weekly active users. Built from event 7. What you decide with it: is usage growing, flat or falling?
• Retention by weekly cohort. Built from events 3 and 7. What you decide with it: do users who joined in week 1 still come back in week 4?
• Paywall conversion. Built from events 10, 11 and 12. What you decide with it: change price, plan or paywall timing.

The version for investors

Investors rarely want your whole dashboard. Share screenshots of the live charts. A screenshot from the tool convinces more than numbers typed into a slide. For more on what to show before a raise, see our investor and fundraising articles. They want three or four lines they can trust:

• 1. Signups per week since launch.
• 2. Activation rate: the share of signups who completed the core action.
• 3. Week 4 retention for your first cohort.
• 4. Paying users and revenue, if you charge.

Your first 30 days, week by week

Follow this order, one funnel at a time. For what comes after the first month, browse our post-launch guides.

• Launch day. Open the app on web, iOS and Android. Sign up, go through onboarding and pay with a test card. Check that all 12 events show up with their properties. A missing event caught on day one takes minutes to fix. Caught on day 25, it has cost you a month of data.
• Week 1. Watch only the signup funnel, because retention means nothing yet. Fix the biggest drop you see.
• Week 2. Switch to the activation funnel. Find the onboarding step where most users leave, then shorten it, remove it or move it after the core action.
• Week 3. Look at feature_used and list the features nobody opens. Talk to five active users and ask what they came for.
• Week 4. Read your first retention cohort and compare it with your second. If the newer cohort holds better, your fixes worked. After one month, that comparison is the strongest story you can tell an investor.

How we handle it at BuildMVPFast

We build MVPs for web, iOS and Android. Fixed at $3,460, delivered in around 20 days, with progress shown every week. The code is 100% yours, on your GitHub, and you get 2 weeks of fixes after launch. The MVP Development page lists what the package covers. You can use this plan with any developer, or talk it through on a free 30-minute call. Want your MVP live and ready to show investors? Book a call.

What is the most important event to track in an MVP?

core_action_completed. It marks the first time a user gets the result your app promises. Signups show interest, and the core action shows value. Your activation rate is based on it.

How many events should an MVP track?

Start with around 12 named events and a few user properties. That covers acquisition, activation, engagement and revenue. Add a new event only when you have a question the current ones can't answer.

Which analytics tool should I use for my MVP?

Use one tool that works on all your platforms and that your developer already knows. PostHog, Mixpanel, Amplitude and Firebase Analytics can all build the five charts in this guide. The event plan matters more than the tool.

When can I trust my retention numbers?

Wait until your first weekly cohort has been live for at least 4 weeks. Before that, you only have a few days of data per user. In weeks 1 and 2, focus on the signup and activation funnels.