App monetization metrics that matter: eCPM, ARPDAU and retention

App monetization dashboards are full of numbers, and it is easy to chase the one that moves fastest. The trouble is that the most visible metric, eCPM, can rise while the business gets worse. This post defines the three metrics that matter most for ad-supported apps, shows how they affect each other, and explains what to optimize and in what order.

eCPM: the price of your inventory

eCPM stands for effective cost per mille, or effective revenue per thousand impressions. It lets you compare the value of inventory across ad networks, formats and pricing models on a single scale.

Formula: eCPM = (ad revenue ÷ impressions) × 1,000

eCPM is useful for comparing demand sources and formats. It tells you how well each impression is being priced. What it does not tell you is how many impressions you are generating, whether users saw ads they wanted to see, or whether those ads pushed people out of the app.

Keep a few caveats in mind:

• eCPM varies naturally by country, platform, format and season, so compare like with like.
• Rewarded and interstitial formats usually price higher than banners, which can make a format mix look better or worse than it really is.
• Cutting low-value impressions raises average eCPM even if total revenue falls.

That last point is why eCPM should never be the only goal.

ARPDAU: revenue per active user

ARPDAU stands for average revenue per daily active user. It connects ad performance to the size of your audience.

Formula: ARPDAU = total daily revenue ÷ daily active users (DAU)

You can also think of it as the product of three levers:

ARPDAU = (ad requests per DAU × fill rate × eCPM) ÷ 1,000 (for ad revenue)

Here, ad requests per DAU reflects how many ad opportunities each user generates, and fill rate is the share of ad requests that return an ad. Multiplying the two gives impressions per DAU. Breaking ARPDAU down this way shows where a change came from. If ARPDAU drops, you can check whether users saw fewer ads, fewer requests were filled, or prices fell.

ARPDAU is more honest than eCPM because it rewards total value per user rather than the price of individual impressions. If you include in-app purchase revenue alongside ads, it also shows how the two models work together.

Retention: the metric that multiplies everything

Retention measures how many users come back after they first open the app. It is commonly reported as day 1, day 7 and day 30 retention.

Formula: Day N retention = users active on day N after install ÷ users who installed on day 0

Retention matters because revenue is earned over a user’s lifetime, not in a single session. A user who returns for weeks generates far more impressions than one who leaves after the first day. This is where lifetime value (LTV) comes in:

LTV ≈ ARPDAU × average number of days a user stays active

The implication is simple. Any change that lifts ARPDAU but shortens how long users stay can reduce lifetime value. Retention is the multiplier on everything else you do.

How the metrics interact

These three metrics pull against each other in predictable ways:

1. More ads per session raise impressions per DAU and short-term ARPDAU, but too many interruptions can hurt retention.
2. Higher floors or stricter demand filters raise eCPM but can lower fill rate and total revenue.
3. User-first formats such as opt-in rewarded video often support retention because users choose to engage, while poorly timed interstitials tend to do the opposite.

Mediation adds another layer. Mediation is the system an app uses to choose among multiple ad sources for each opportunity. In a traditional waterfall, networks are called one after another in a fixed order based on historical eCPM, and the first one to fill wins. In-app bidding replaces much of that with a real-time auction where sources bid at the same time, so the highest price can win regardless of order. Waterfalls can leave money on the table when a lower-ranked network would have paid more, which is one reason many apps now use bidding or a hybrid of the two.

When you review mediation performance, look at revenue and ARPDAU by source, not just eCPM. A network with a high eCPM and low fill may contribute less than one with a moderate eCPM and consistent fill.

What to optimize, and in what order

A practical sequence for most ad-supported apps:

• Protect retention first. Track day 1 and day 7 retention alongside every monetization change. If a test lifts revenue but hurts retention, treat it as a failure until proven otherwise.
• Optimize ARPDAU second. Use it as the main revenue goal, broken into impressions per DAU, fill rate and eCPM so you know which lever moved.
• Use eCPM as a diagnostic. It is best for comparing demand sources, formats and regions, not as a headline target.
• Test in cohorts. Compare groups of users who installed at the same time and saw different ad settings, and follow them long enough to see retention effects.

Vortex is growing its work with app publishers, connecting in-app inventory to curated demand with formats designed for user trust and retention. If you want a second opinion on your app’s monetization setup, talk to the Vortex team.