Data is the core of user acquisition. Every decision — from which ad to pause, to how much budget to shift, to whether a campaign is worth scaling — should be backed by real numbers.
Performance data comes from two main sources: the ad platforms themselves (Meta Ads Manager, TikTok, Google UAC), which show campaign-level data like impressions, clicks, and installs; and analytics tools like MMPs and product dashboards, which reveal deeper signals about user quality and revenue.
Understanding both is what gives you the full picture.
Inside ad platforms, you see metrics from the ad impression through to the app install, and sometimes early in-app events if tracking is properly configured. These help you evaluate how efficiently you're acquiring users.
Here are the core metrics you'll work with daily:
Cost per Install (CPI) tells you how much you're paying per app install. Formula: Total Ad Spend ÷ Number of Installs. This is your baseline efficiency metric.
Cost per Action (CPA) measures the cost to acquire one user who completes a target action — starting a trial, making a purchase, subscribing. Formula: Total Ad Spend ÷ Number of Target Actions.
Click-Through Rate (CTR) shows how engaging your ad is — what percentage of people who see it actually click. Formula: (Clicks ÷ Impressions) × 100%.
Cost per Click (CPC) is your cost efficiency at the click level. Formula: Total Ad Spend ÷ Clicks.
Cost per Mille (CPM) reflects auction cost — what you're paying per 1,000 impressions. Formula: (Total Ad Spend ÷ Impressions) × 1,000. This is driven by competition and platform demand, not something you directly control.
The value of these metrics is in how they connect. If your trial costs suddenly spike, you work backward through the funnel. Maybe CTR dropped, meaning the creative isn't resonating. Maybe CPMs rose because of seasonal competition. Maybe a strong-performing creative got paused by mistake. Funnel-based analysis helps you troubleshoot step by step rather than guessing.
After the install, the real evaluation begins — did the user sign up, start a trial, subscribe, or come back?
Some post-install data can be tracked directly in ad platforms using event postbacks. But for deeper analysis, you typically need MMPs (Mobile Measurement Partners) like AppsFlyer, Adjust, or Branch.
MMPs help with two critical things. First, attribution — determining which platform actually brought the user and avoiding double-counting across channels. Second, deeper analytics that ad platforms can't provide on their own.
Here are the key post-install metrics:
ROAS (Return on Ad Spend) shows how much revenue your ads bring in relative to cost. Formula: Revenue from Ads ÷ Total Ad Spend. This is your north star for profitability.
LTV (Lifetime Value) measures how much value an average user generates over time. Typically calculated as average revenue per user over a defined window (D7, D30, D90).
CAC (Customer Acquisition Cost) is the cost to acquire one paying customer. Formula: Total Marketing Spend ÷ Paying Users. This often equals CPA but can include broader marketing costs.
Retention (D1/D7/D30) tracks how many users keep using the app after install. Formula: (Users Returning on Day X ÷ Installs on Day 0) × 100%. This is the clearest signal of whether you're acquiring the right users.
ARPU and ARPPU — Average Revenue Per User includes everyone (free and paid), while Average Revenue Per Paying User focuses only on monetizing users. Both help you understand revenue dynamics.
Subscription Conversion Rate measures how many installs or trials become paying subscribers. Formula: (Subscribers ÷ Installs or Trials) × 100%.
Payback Period tells you how long it takes to earn back your CPI or CAC. The faster the payback, the more aggressively you can reinvest in acquisition.
For easy reference, here's how these metrics group:
| Category | Metrics |
|---|---|
| Core UA | CPI, CPA, ROAS, CAC |
| Creative | CTR, CPC, CPM |
| Monetization | LTV, ARPU, ARPPU, Subscription Conversion Rate |
| Engagement & Retention | D1/D7/D30 Retention, DAU/MAU, Session Frequency |
The goal isn't to track everything all the time — it's to know which metrics matter for the decision you're making right now, and how they connect to the business outcome you're driving toward.
Need help with paid acquisition for your app?
Book a call →Meta advertising looks meaningfully different in 2026 than it did two years ago. Here's what's actually working for digital product campaigns — from Andromeda to Advantage+ to attribution changes.
Most teams that try TikTok ads fail for the same reason — they run it like Meta. Here's what's actually different about the platform and what changes when you get it right.