ARPU calculation explained: the formula, a step-by-step method, variants by business model and benchmark levels from 147 B2C operators.
Frequently asked
Related resources
Solutions
See it in the product
The same loops this note describes run 24/7 against your customer base. Watch the workspace decide, experiment and execute 1:1.
Vocabulary
Definitions an answer engine can quote, each one a page of its own.
Compare
Neutral reads on the categories buyers evaluate against a decision layer.
All comparisonsARPU stands for Average Revenue Per User. It is the revenue a business earns, on average, per active customer over a defined period. The ARPU calculation is simple, revenue divided by users, but the two inputs are where almost every reporting error lives. This guide covers the formula, a step-by-step calculation, the variant each business model needs, and benchmark levels from 147 B2C operators so you can place your own number.
Both terms must cover the same period and the same population. A quarterly revenue number divided by a monthly user count is not a real quantity. Neither is a revenue figure that includes professional services divided by a user count that only counts end consumers. This is boring to say and universally violated.
To calculate ARPU: pick one period, usually a calendar month. Sum recognized revenue attributable to end users in that period, net of refunds, chargebacks and taxes. Count the users who were active in the same period under one fixed definition. Divide revenue by users. Report the definition alongside the number.
A streaming service recognizes $12,000,000 in subscription revenue in July, net of refunds and VAT. It opens the month with 3,900,000 paying subscribers and closes with 4,100,000, so average active users are 4,000,000. Monthly ARPU is $12,000,000 / 4,000,000 = $3.00. Had the team used the closing count instead, ARPU would read $2.93, a 2.4% understatement produced entirely by the denominator.
The formula is constant; the numerator and denominator are not. Four common cases, with the trap in each:
| Model | Numerator | Denominator | Trap |
|---|---|---|---|
| Subscription | Net recognized subscription revenue | Average paying subscriptions in the period | Counting paused and grace-period accounts as active |
| Prepaid / consumption | Net revenue from consumed balance, not top-ups sold | Users with at least one revenue event in the period | Recognizing top-ups on sale, which front-loads ARPU |
| Freemium with ads | Subscription plus net advertising revenue | All active users, paying and free (use ARPPU separately) | Reporting ARPPU as ARPU, inflating the figure several times over |
| Marketplace | Take rate on GMV, not GMV itself | Active buyers in the period, stated as buyers not accounts | Dividing by registered accounts instead of transacting buyers |
Monthly ARPU is the operating number. Annual ARPU is not twelve times monthly ARPU unless the base is flat: multiplying a monthly figure by twelve on a growing base inflates the result, because the later months carry more users at a lower average. Annualize by summing twelve months of revenue and dividing by the twelve-month average user count.
Cohort ARPU, revenue per user for customers acquired in the same month, tracked by tenure, is the version that actually informs decisions. Blended ARPU moves when the mix of new and mature customers moves, which means a healthy acquisition quarter can look like an ARPU decline. Cohort ARPU separates the two.
There is no universal threshold, but there are category anchors. Across 147 B2C operators with at least one million active customers, median monthly ARPU runs from roughly $4 in prepaid telecom to roughly $35 in postpaid telecom, and the top quintile sits 45% to 150% above the median inside the same category. Window: TTM to Q2 2026.
| Industry | n | Region | Median ARPU / mo | Top-quintile ARPU / mo |
|---|---|---|---|---|
| Streaming | 18 | US / EU | $9.40 | $14.80 |
| Telecom (postpaid) | 24 | US / EU | $34.60 | $48.20 |
| Telecom (prepaid) | 21 | LATAM / SEA | $4.10 | $7.30 |
| Fintech | 19 | US / EU | $6.80 | $13.10 |
| Travel | 12 | Global | $11.20 | $19.60 |
| Marketplaces | 16 | Global | $7.50 | $15.40 |
| Gaming (live services) | 14 | Global | $5.20 | $12.90 |
| Media / publishing | 13 | US / EU | $8.10 | $13.70 |
| Retail / subscription commerce | 10 | US / EU | $12.60 | $22.40 |
Source: Markin B2C ARPU benchmark set, TTM to Q2 2026. Levels are monthly ARPU in USD at period-average rates and are not comparable across geographies. Full method, growth rates and disclosure basis per row in ARPU benchmarks by industry 2026.
Three acronyms show up interchangeably and mean subtly different things:
None of these is more correct than the others; they are different lenses on the same base. What matters is disclosing which one you use and not switching definitions between periods without saying so.
Four errors account for most of the ARPU discrepancies we see when auditing a client's reporting. The magnitudes below are the typical distortion on a base of one million-plus customers.
ARPU only carries information when compared inside a category, controlling for tenure, plan mix and geography. A prepaid mobile carrier in Southeast Asia and a US streaming service can have identical growth stories with 20x different ARPU. Absolute levels across categories tell you almost nothing.
ARPU is a per-period rate. LTV (Lifetime Value) is the discounted sum of future ARPU over the expected customer lifetime, net of variable cost. LTV depends on both ARPU and retention; ARPU depends on neither. A cohort with high ARPU and high churn can have lower LTV than a cohort with lower ARPU and long tenure.
ARPU moves for many reasons that have nothing to do with any program: seasonality, mix changes, tariff changes, currency movements, cohort maturation. Attributing quarter-over-quarter ARPU deltas to a specific campaign or a specific product change without a preserved holdout is almost always overclaiming.
A defensible ARPU report has: (1) an explicit revenue definition (net, gross, tax treatment); (2) an explicit denominator definition (paying, active, entitled); (3) a breakdown by tenure cohort and plan tier so mix effects are visible; and (4) an incremental-lift number against a preserved holdout wherever ARPU is being actively influenced by a growth program.
The last point is where most operators leave the largest ARPU gains unrealized. The mechanism that closes it is per-customer decisioning measured against a control, see next best action marketing. For a practical framework on turning ARPU into a growth rate rather than a reporting metric, see how to increase ARPU: a framework for B2C enterprises. For where the leaders sit today, see ARPU benchmarks 2026.
The benchmark levels on this page come from the Markin B2C ARPU benchmark set: 147 operators with at least one million active customers and 24 months of ARPU history, window TTM to Q2 2026. ARPU is recognized monthly revenue attributable to end users, net of refunds and taxes, over average active users. Some rows are anchored on publicly disclosed ARPU, others derived from observed data; the per-row basis is published in the benchmarks post. Levels are not comparable across geographies, and a single-country operator should read its own region's rows only.
Markin measures ARPU as incremental lift against a preserved holdout on your own base. To see the number your growth program is actually earning, explore Revenue discovery.
