Calculate Ad Revenue
Learn how to accurately calculate ad revenue by understanding the complexities behind the dashboard figures.

You open AdMob, AppLovin, or your mediation dashboard and the graph looks healthy. Impressions are flowing. eCPM doesn't look bad. Then you check what arrived in your account, and the number feels smaller than the dashboard implied.

That gap frustrates almost every app founder at some point. Usually, the problem isn't that the dashboard is lying. It's that the dashboard is showing one layer of the truth, while your business runs on several layers at once.

If you want to calculate ad revenue properly, you need more than one formula. You need a model that connects impressions, users, session behavior, viewability, mediation logic, and the share you keep. That's when ad revenue stops being a vanity number and starts becoming a diagnostic tool.

Table of Contents

• Beyond the Dashboard Gross Number
  • Why the dashboard feels misleading
  • What a useful revenue model actually does
• The Foundation eCPM and Impression-Based Revenue
  • The simple formula most teams start with
  • What eCPM tells you and what it hides
  • Where impression math is strongest
• Thinking in Users not Impressions with ARPDAU and ARPU
  • Why the denominator changes the story
  • The two user metrics that matter most
  • When to prioritize user metrics over eCPM
• Accounting for Mediation and Revenue Share Realities
  • Why gross numbers overstate what you keep
  • What to look for in your own stack
  • The founder's mental model
• Worked Example Building Your Revenue Spreadsheet
  • The PhotoSort AI example
  • The core calculations
  • What this spreadsheet should reveal
  • The practical build standard
• Common Pitfalls and How to Increase Your Revenue
  • The mistakes that distort revenue analysis
  • What usually works better
  • The levers worth testing
  • The human judgment layer

Beyond the Dashboard Gross Number

The biggest mistake I see is treating the top-line dashboard number as the business. It isn't. It's a surface metric.

A monetized app has multiple translation steps between user activity and cash. A user opens the app. That user creates sessions. Sessions create ad opportunities. Some of those opportunities become viewable impressions. Some of those impressions get filled at a useful price. Then networks, mediation layers, and payout rules shape what you keep.

Why the dashboard feels misleading

A blended revenue graph hides the mechanics underneath it. That's why founders often say something like, “Traffic is up, so why doesn't revenue feel up?” The answer is usually buried in user behavior, placement quality, or inventory quality.

A dashboard can look strong while the underlying monetization is weak if:

• Users aren't generating enough sessions: Installs happened, but engagement is shallow.
• Placements exist but aren't high quality: Ads load, but they don't become valuable impressions consistently.
• Geography mix changes: More users can still mean lower monetization if the traffic mix shifts.
• One network props up the average: A healthy blended eCPM can hide weak performance almost everywhere else.

Practical rule: Don't ask “How much did ads make?” first. Ask “Which user actions created the revenue?”

What a useful revenue model actually does

A good model lets you answer operational questions, not just accounting questions.

For example:

That's the reason to calculate ad revenue carefully. The formulas tell you what your users are doing, how your ad stack is interpreting that behavior, and where your monetization system is losing efficiency.

The Foundation eCPM and Impression-Based Revenue

If you want the simplest way to calculate ad revenue, start with impressions and eCPM.

An impression is one ad view. eCPM means effective cost per mille, or the effective revenue per thousand impressions. The basic formula is straightforward:

Revenue = (Impressions / 1,000) × eCPM

That's the cleanest top-line estimate for ad-supported inventory.

!A diagram explaining the formula for calculating ad revenue based on impressions and eCPM.

The simple formula most teams start with

In practice, publishers often begin with CPM-based math because earnings are tied to impressions rather than installs or clicks. Google News Initiative's framework explains the underlying mechanics as viewable impressions = ad units × pageviews × viewability, then divide by 1,000 and multiply by CPM and sell-through rate. The same explanation gives a concrete example: 100,000 pageviews at a $10 CPM yields $1,000 in revenue math for that traffic volume, as summarized in Taboola's write-up of the Google News Initiative formula.

That matters because the visible revenue number doesn't start with money. It starts with inventory creation.

What eCPM tells you and what it hides

eCPM is useful because it compresses a messy system into one comparable rate. You can look at one ad unit, one country, one format, or one day and ask, “How much value did each thousand impressions produce?”

But eCPM is also a blended number. That's where founders get into trouble.

If one rewarded placement performs extremely well and banners perform poorly, your app-level eCPM can still look decent. If one country monetizes strongly and another doesn't, the average can mask the problem. If viewability is weak, a CPM estimate can make the business look healthier than it is.

A high eCPM doesn't automatically mean a healthy ad business. Sometimes it just means a small slice of inventory is carrying the average.

Where impression math is strongest

Use impression-based revenue when you want a fast estimate of monetization output from the inventory itself. It's especially useful for:

• Placement analysis: Compare interstitials, banners, native, and rewarded units.
• Network comparison: See which demand sources monetize inventory better.
• Creative format decisions: Different formats often change impression quality.
• Unit economics by ad slot: At this stage, a placement review, including formats like rich media ads for mobile apps, becomes useful.

Impression math is the foundation. It just isn't the whole building.

Thinking in Users not Impressions with ARPDAU and ARPU

Teams often spend too long staring at inventory metrics and not enough time staring at user economics. That's backwards for app monetization.

Impressions tell you what got served. ARPDAU and ARPU tell you what a user is worth. For product, retention, and acquisition decisions, that's often the more important lens.

!A professional analyzing a user-centric revenue dashboard on a digital tablet in a modern office workspace.

Why the denominator changes the story

Google's training materials make a point that most pages about how to calculate ad revenue don't answer the harder question: which traffic mix makes the estimate break down? They also note that many guides isolate CPM, RPM, sessions, page views, CTR, CPC, and ARPU without explaining when each denominator fits the business model. The useful contrarian framing is that revenue per 1,000 is often less useful than revenue per session or revenue per user in app monetization, as explained in Google News Initiative's ad revenue estimation training.

That's exactly right for apps.

If two apps produce the same impression volume, they can still have completely different economics. One may rely on a small number of heavy users with deep session behavior. The other may have broad but shallow traffic that looks good in aggregate and weak in retention.

The two user metrics that matter most

Use these formulas:

• ARPDAU = Daily ad revenue / Daily active users
• ARPU = Revenue over a period / Total users in that period

They sound similar, but they answer different questions.

ARPDAU is usually the sharper tool for ad-supported apps because it ties monetization directly to active engagement. If you change ad placements, frequency, or rewarded flows, ARPDAU tells you whether those active users are producing more revenue.

ARPU matters when you care about the broader relationship between acquisition, monetization, and retention across the whole user base.

When to prioritize user metrics over eCPM

At this stage, a lot of teams mature. They stop obsessing over whether eCPM is “good” and start asking whether users are becoming more valuable.

Prioritize ARPDAU or ARPU when:

• You're evaluating acquisition traffic quality
• You're comparing countries with different session depth
• You're testing product changes that alter user behavior
• You're deciding whether ad load is hurting retention

If eCPM rises while ARPDAU stays flat, users probably aren't generating more monetizable behavior. The ad stack got more efficient, but the product didn't.

That distinction matters because founders often over-credit monetization changes for gains that really came from traffic mix, seasonality, or one high-performing segment. User-based metrics force cleaner thinking.

Accounting for Mediation and Revenue Share Realities

Gross ad revenue is not net ad revenue. That sounds obvious, but many spreadsheets still stop too early.

Your app usually doesn't sell every impression directly to advertisers. Mediation platforms route opportunities across multiple networks. Networks apply their own economics. Payout timing and reporting logic create another layer of distance between impression value and cash received.

Why gross numbers overstate what you keep

A dashboard may show estimated earnings at the network or mediation level, but your business needs to understand the chain:

1. Users create ad opportunities
2. Those opportunities enter a mediation or network stack
3. Demand sources bid or fill
4. The platform reports revenue
5. The payout you receive reflects the platform's terms

Each layer can reduce what feels like “obvious” revenue in the first-pass math.

A blended estimate based only on impressions and eCPM is still useful, but it's pre-reality. You haven't yet accounted for how the inventory was distributed, which buyers won, whether one platform favored its own demand, or what share of gross value flowed through to you.

What to look for in your own stack

Don't guess. Pull the exact reporting definitions from the tools you use.

Review these items inside your monetization setup:

• Mediation reporting rules: Some platforms report estimated revenue in ways that differ from final payout accounting.
• Network-by-network payout logic: Not every demand source represents revenue the same way.
• Waterfall or bidding behavior: The structure affects which impressions go where.
• Final remittance reports: Finance and growth need to reconcile the numbers using these reports.

A practical revenue model should include both a gross estimate and a realized revenue view. If those numbers are consistently far apart, that's not a spreadsheet issue. It's a signal that your stack needs auditing.

The founder's mental model

Think of mediation as a routing system, not a magic revenue multiplier. It can improve monetization quality, but it also adds abstraction.

The operators who calculate ad revenue well don't stop at one total. They break it into layers:

That separation prevents bad decisions. Without it, founders can increase ad load, hurt user experience, and still fail to improve the number that matters.

Worked Example Building Your Revenue Spreadsheet

A revenue spreadsheet should be boring. If it feels clever, it's probably too fragile.

Use one sheet for raw inputs, one for calculated outputs, and one for notes about reporting definitions. The notes matter more than people think because different platforms label revenue in slightly different ways.

!A five-step infographic showing the process of building a revenue spreadsheet for an AI photo app.

The PhotoSort AI example

Take a fictional app called PhotoSort AI. It monetizes with ads and uses a mediation layer across three networks.

You want one daily sheet with these columns:

The exact values will be your own. The important part is structure.

The core calculations

The spreadsheet logic should flow like this:

1. Calculate total impressions
2. Allocate impressions by network
3. Apply each network's eCPM
4. Sum gross revenue
5. Compare reported revenue against realized revenue

For each network, the formula is:

Network revenue = (Network impressions / 1,000) × Network eCPM

Then sum the network rows for total gross revenue.

Here's a stripped-down mock layout:

Once the gross total is there, add a second section below it for the finance-facing view:

• Estimated gross revenue
• Reported platform revenue
• Payout received
• Variance notes

That final row is where teams usually learn the most.

Keep a comments column for every manual assumption. Six weeks later, nobody remembers why one day looked different.

What this spreadsheet should reveal

A good spreadsheet doesn't just total revenue. It surfaces patterns.

You should be able to answer questions like:

• Which network carries the most value?
• Did a change in DAU improve monetization, or just increase raw traffic?
• Did more impressions per user help, or did they just dilute user experience?
• Is one country, source, or placement distorting the average?

If you want to make it more useful, add tabs for cohorts or geographies. Don't add complexity for its own sake. Add it only when it helps you isolate why revenue moved.

The practical build standard

The best version of this sheet is usually not the prettiest one. It's the one a growth manager, founder, and finance lead can all read without asking for translation.

Use these build rules:

• One metric per column: Don't combine raw inputs and calculated outputs in the same field.
• One definition per tab: If DAU comes from one analytics source, document it once and keep it consistent.
• One time zone: Mixed time zones create fake revenue problems.
• One reconciliation routine: Match dashboard estimates against actual payout on a regular cadence.

That's how you calculate ad revenue in a way that survives real operations instead of collapsing under dashboard screenshots.

Common Pitfalls and How to Increase Your Revenue

Most revenue problems aren't caused by one catastrophic issue. They come from small modeling mistakes and weak prioritization.

Teams focus on the metric that looks easiest to improve, usually blended eCPM, and ignore the metric that would tell them whether the business got healthier. That's how an app can “optimize monetization” and still make poor growth decisions.

!An infographic showing common ad revenue pitfalls versus strategies to increase mobile advertising earnings and growth.

The mistakes that distort revenue analysis

The first trap is relying on one blended app-wide number. That average is too polite. It hides bad placements, weak geographies, low-value users, and underperforming traffic sources.

The second trap is optimizing ad delivery without protecting user behavior. An app can squeeze more impressions out of a session while making the product worse. If retention softens, that short-term gain becomes a longer-term tax.

The third trap is failing to segment. If you don't separate revenue by placement, country, source, and active-user behavior, you can't tell whether monetization improved or whether the traffic mix changed.

What usually works better

Use a narrower operating view.

• Break out monetization by placement: Rewarded, interstitial, banner, and native inventory behave differently.
• Review by geography and traffic source: A rise in installs can still drag revenue quality down.
• Judge changes with user metrics too: ARPDAU often catches weak monetization decisions faster than blended eCPM.
• Treat ad load as a product decision: The best monetization changes preserve the user loop that creates future sessions.

Better monetization usually comes from better alignment between user behavior and ad opportunity, not from forcing more ad pressure everywhere.

The levers worth testing

Not every optimization requires a full stack rebuild. In many apps, the highest-impact work is operational.

For broader context, scale amplifies these decisions dramatically. Statista reports that Meta generated over 196 billion U.S. dollars in advertising revenue in 2025, up 22.1% from the previous year, which is why standardized monetization metrics matter so much when evaluating efficiency at any scale, from a single app to a global platform, as shown in Statista's Meta advertising revenue dataset.

That same logic applies to app publishers. Small improvements in monetization efficiency can matter a lot when repeated across large impression volume, deep session loops, or broad user bases. If you're refining your acquisition-to-monetization system, it helps to understand the wider mechanics of mobile app advertising strategy.

The human judgment layer

AI helps, but doesn't replace operators. AI can speed up reporting, spot patterns, and help generate test ideas. It can't decide which monetization trade-off is acceptable for your product.

Founders still need to answer the hard questions:

• Which placements fit the product experience?
• Which users should see more ads, and which should see fewer?
• When is a monetization gain too expensive in retention terms?
• Which metric reflects the actual health of the app right now?

The future of ads will reward teams that combine faster execution with sharper judgment. That's true in acquisition, and it's just as true when you calculate ad revenue.

If your app is getting traffic but monetization still feels underwhelming, Marketing For Apps By @designerants helps mobile app teams improve the part most companies underinvest in: the ads themselves. Better copy, clearer hooks, stronger creative angles, and ads that generate desire can improve the quality of the users you acquire, which gives every downstream monetization metric a better starting point.