Unit Economics and Revenue Metrics for Games and Apps

You can have a million downloads and still lose money. Scale does not fix a broken model; it multiplies whatever your unit economics already say, good or bad. If each user costs more to acquire than they return, growth just makes the hole deeper.

This article looks at unit economics and revenue metrics from the angle that matters most for app and game teams: the loop between what you spend to get a user and what that user pays back. The aim is working math that survives past launch week.

What unit economics actually measures

Unit economics compares the revenue from a single user against the cost of getting and serving that user. For games and apps, a unit usually means one install or one paying customer. The core figure is lifetime value (LTV) against customer acquisition cost (CAC).

LTV depends heavily on assumptions about retention and purchase behavior. Two analysts can run the same game and get wildly different LTV numbers depending on how many months of data they include. CAC is cleaner, but it hides channel mix, ad spend on organic installs, and refunds. Treat both numbers as estimates you update, not fixed truths.

Calculating unit economics means working from a single customer or transaction outward to the numbers that decide whether growth creates or destroys value. The sequence below follows that logic, starting with the contribution margin per unit and ending with the revenue metrics that show whether the model holds as volume increases.

A step-by-step pass through the model

1. Define the unit precisely: one install, one subscription account, or one active seat. State which costs are direct (payment processing, hosting per account, support tickets attributable to that unit) versus shared overhead that stays out of the calculation.
2. Compute contribution margin per unit as revenue per unit minus variable cost per unit. Compare it against customer acquisition cost (total sales and marketing spend in a period divided by new units acquired in that same period) to get the CAC payback period in months.
3. Set the churn and expansion inputs: monthly churn, gross revenue churn, and expansion revenue from upgrades or added seats. Net revenue retention depends on the gap between the last two.
4. Derive lifetime value from contribution margin divided by revenue churn. Test the result under a pessimistic churn assumption rather than only the trailing average.
5. Reconcile the unit-level view to reported revenue: bookings, recognized revenue under the applicable accounting standard, and deferred revenue for prepaid or annual contracts. The bottom-up model should tie to the financial statements.
6. Rerun the model by cohort and by acquisition channel. Blended averages hide channels whose payback exceeds the cash runway or cohorts that never reach the retention assumed in step four.

The output of this sequence is a working model, not a single number. The assumptions in steps one and three drive most of the variance, so document them and revisit them each period. Where a metric cannot be computed from available data, such as expansion revenue for a product without tiered pricing, record it as not applicable rather than substituting a proxy that will later be quoted as fact.

The revenue metrics that feed LTV

LTV is a summary. The metrics underneath it are where decisions get made. If you only track one blended LTV number, you cannot tell whether a monetization change helped, hurt, or simply moved spending around.

A handful of revenue metrics drive most of the story for games and apps:

• ARPU (average revenue per user): total revenue divided by active users, useful for a broad view.
• ARPPU (average revenue per paying user): revenue divided by payers only, which shows monetization depth.
• Payer conversion rate: the share of active users who make at least one purchase.
• Retention by day (D1, D7, D30): the curve that decides how long users keep generating revenue.
• Repeat purchase rate: how often payers come back and buy again, especially for consumables and battle passes.

Together these metrics explain why LTV moved, not just that it moved. ARPPU and repeat purchase rate show monetization depth, while payer conversion and retention show how many users reach a paying moment at all. Track them side by side and a rising LTV stops being ambiguous.

ARPU can rise while your business worsens. Suppose you lose your light users and keep only whales. ARPU climbs and payer conversion looks decent, but your addressable market shrinks. Pair every average with a distribution view so a shrinking user base does not masquerade as improving monetization.

Matching metrics to your monetization model

Not every model produces revenue the same way, so the metrics that matter shift. A subscription app cares about churn and expansion. An ad-supported game cares about impressions per session and fill rate. A free-to-play game with in-app purchases cares about payer depth and repeat purchases.

This table maps each model to the metrics worth watching most closely.

The table is not a ranking. A model with a high ARPPU can still have terrible unit economics if CAC is high or retention collapses early. The right model is the one whose revenue curve you can measure honestly and forecast with some confidence.

Payback period tells you more than LTV/CAC

Many teams compare LTV and CAC as a single ratio, often aiming for something like 3:1. That ratio is useful, but payback period answers a more practical question: how many days until a user has returned their acquisition cost?

Short payback matters for cash flow. A game that recovers CAC in weeks can reinvest in growth quickly. One that takes years needs patient capital and carries more risk, because retention assumptions can break over that horizon. Payback period forces you to talk about time, and time is where most monetization plans fall apart. A 3:1 LTV/CAC spread over four years is far weaker than the same ratio recovered inside a quarter.

Where the payments layer changes the numbers

Many teams overlook a quiet adjustment in unit economics: the revenue you count per user is usually gross. What you actually keep is net of store fees, payment processing, tax, and refunds. That gap matters when margins are thin.

Consider a subscription that renews monthly. Each renewal carries processing cost and, in some stores, platform commission. If refunds run above your assumption, net revenue per user drops and your payback period stretches. Track net revenue alongside gross, and reconcile the two. A monetization model that looks healthy on gross revenue can be fragile once fees and refunds are subtracted. Regional pricing and tax rules can widen the gap between your headline ARPU and what lands in the bank.

Building the number you can defend

The goal is not a perfect model. It is a model whose inputs you can explain and update. Start with cohort-level data: how a group of users retained, converted, and paid over time. Cohort data removes the illusion created by blended averages.

From there, set a target payback period, measure net revenue, and watch retention closely. Review the assumptions behind LTV on a fixed schedule, especially after a monetization change or a shift in ad spend. If a metric moves, know which input caused it before you touch the model.

Unit economics and revenue metrics will not tell you whether a game is fun or an app is useful. They tell you whether the business behind it can keep going. That is the question worth answering before you scale anything.