CCalclabhub

Business · 44 calculations

Metrics

How To Calculate ARPU

ARPU tells you how much revenue each active customer contributes over a set period. It distils a whole user base into one comparable number, and it is the starting point for annualised figures, per-thousand rescalings and daily rates.

Last updated

Quick Answer

ARPU = Total Revenue / Active Users

ARPU
Average revenue per active user for the period
Total Revenue
All revenue booked in the measured period
Active Users
Number of users counted as active in the same period

Divide total revenue for a period by the number of active users in that same period. With 1,200,000 in monthly revenue and 50,000 active users, ARPU is 24.0. Multiply by 12 to annualise (288.0), scale to a thousand users (24000.0), or divide by 30 for a daily rate (0.8).

What Is ARPU?

Average revenue per user, usually shortened to ARPU, is total revenue divided by the number of active users over a defined period. If a service books 1,200,000 in a month from 50,000 active customers, each customer contributed an average of 24.0 for that month. The figure is an average in the strict sense: it does not claim that any particular customer paid exactly that amount, only that the total spread evenly across the base gives that value. It is one of the most widely quoted metrics in subscription businesses because it reduces a complicated revenue stream to a single, easily communicated number. Because it is a ratio, ARPU can be computed for any grouping of users, from a single country to an entire product, simply by choosing a matching revenue and user count for that group.

The metric earns its place because it lets you compare things that would otherwise be incomparable. A small app with 5,000 users and a large platform with 5,000,000 users cannot be compared by total revenue alone, but their ARPUs sit on the same scale. Analysts use it to compare months, to compare acquisition cohorts, to contrast regions, and to track whether a product is monetising its audience better over time. Rising ARPU with a flat user base means each customer is worth more; falling ARPU with a growing base can mean cheaper acquisition is bringing in lower-value users. Tracking the same cohort month by month is especially revealing, because it shows whether individual customers naturally spend more over time or whether growth comes only from replacing them with wealthier newcomers.

Getting the numerator right matters as much as the division. Total revenue should be the amount actually recognised for the period, net of refunds, discounts and credits where that is the convention your team uses. Mixing gross bookings in one period with net revenue in another creates a false trend. One-off items such as a large contract settlement can distort a single month badly, which is why many teams report both a blended figure that includes everything and a recurring-only figure that strips such events out.

The denominator is where most of the real disagreement lives. There is no universal definition of an active user: some businesses count anyone who logged in during the month, others only paying subscribers, and others anyone who completed a meaningful action. Each choice produces a different ARPU from the same revenue. What matters is consistency, so that a change in the metric reflects the business changing rather than the definition changing. The related measure ARPPU deliberately divides by paying users only, which strips out free or inactive accounts and therefore always gives a larger number.

Annualising turns a short-period figure into a yearly one. The calculation assumes the measured period is one month, so the monthly ARPU of 24.0 becomes an annualised 288.0 when multiplied by twelve. This is a projection, not a forecast: it assumes the monthly performance repeats unchanged for a year. If your underlying period is a week you would multiply by 52, and if it is already a year there is nothing to annualise at all. Applying the factor of twelve to the wrong period is a common and completely avoidable error. Note also that annualising is a straight multiplication, not a compound one: it assumes no growth in revenue or users across those months, only that the observed month is representative.

Revenue per thousand users is the same ratio rescaled to a convenient block size. Taking 1,200,000 of revenue over 50 users per thousand — because 50,000 users contains fifty groups of a thousand — gives 24000.0 per thousand. This is simply ARPU multiplied by 1,000, and it exists because very small or very large per-user numbers are awkward to read aloud and to compare in a table. Advertising and publishing businesses favour this convention, where revenue per thousand impressions is a familiar unit.

Daily ARPU breaks the monthly figure into a per-day rate by dividing by 30, giving 0.8 from the monthly 24.0. The divisor is a convention: thirty days is a convenient round month, even though real months have 28, 29, 30 or 31 days. Using 365 instead would give a yearly figure divided down to a day, which is a different quantity. State which convention you used, because a daily rate built from a thirty-day month and one built from a calendar year will not quite agree.

ARPU connects directly to customer lifetime value. A simple lifetime value estimate multiplies ARPU by the average number of periods a customer stays active, so a 24.0 monthly ARPU with an average lifetime of twenty months implies roughly 480.0 of revenue per customer. That link is why ARPU movements get so much attention: a small sustained increase compounds across every customer and every month they remain. For the same reason, an ARPU figure that cannot cover the cost of acquiring a customer signals trouble no matter how impressive it looks in isolation. It also explains why a churn problem can quietly destroy value even while ARPU itself looks healthy.

The main limitation is that an average hides its distribution. A base where a handful of heavy spenders generate most of the revenue can show a respectable ARPU while the typical customer pays very little. That makes the mean sensitive to outliers and potentially misleading when guiding product decisions. ARPU is also not profit: it says nothing about the cost of serving a user, acquisition spend or overhead. Used alongside segmented views, retention data and cost metrics it is powerful; used alone it can paint an optimistic and incomplete picture. A generous free tier, for example, can depress the blended figure while the paying minority is highly profitable, so separating free from paying users often tells a very different story.

Formula

ARPU = Total Revenue / Active Users

Formula page →

Divide the revenue recognised in a period by the number of active users in that same period. The two figures must cover identical boundaries or the ratio is meaningless.

SymbolMeaning
RTotal revenue for the period
UActive users in the period

Annual ARPU = ARPU x 12

Formula page →

For a monthly figure, multiplying by twelve projects one month of performance across a year. Change the multiplier if the base period is not a month.

SymbolMeaning
ARPUMonthly average revenue per user
ARPU annualAnnualised average revenue per user

Revenue per 1,000 = Total Revenue / (Active Users / 1,000)

Formula page →

Rescales the same ratio to a block of one thousand users, which is ARPU multiplied by 1,000. Useful when per-user figures are too small to read comfortably.

SymbolMeaning
RTotal revenue for the period
UActive users in the period

How To Calculate ARPU

  1. 1

    Fix the measurement period

    Decide the window you are measuring, such as one calendar month, and make sure the revenue figure and the user count both refer to exactly that window. A mismatch here cannot be repaired later by any arithmetic.

  2. 2

    Count the active users

    Apply your definition of active consistently and count the distinct users who qualify in the period. Avoid double counting accounts, and decide in advance whether free or trial users are included, because every later figure inherits that choice.

  3. 3

    Divide revenue by users

    Take the recognised revenue for the period and divide by the active user count. With 1,200,000 and 50,000 the result is 24.0, the average revenue contributed per active user for that month.

  4. 4

    Annualise and rescale

    Multiply the monthly figure by 12 for an annualised 288.0, divide the revenue by users-per-thousand for 24000.0, or divide the monthly figure by 30 for a daily rate of 0.8. Each is the same underlying ratio expressed on a different scale.

  5. 5

    Interpret and segment

    Compare the result with earlier periods and with cohorts, then check whether the average is being dragged by a few heavy spenders. Segmenting by plan, region or tenure usually reveals more than the blended number alone.

Examples

Example 1: Default case — a month of subscription revenue

totalRevenue
1200000
activeUsers
50000
StepCalculationResult
Base ARPU1200000 / 5000024.0
Annualised ARPU24.0 x 12288.0
Revenue per thousand users1200000 / (50000 / 1000)24000.0
Daily ARPU24.0 / 300.8

Result: With 1,200,000 in revenue and 50,000 active users, monthly ARPU is 24.0, annualised ARPU is 288.0, revenue per thousand users is 24000.0, and daily ARPU is 0.8.

Example 2: A smaller app — rescaling to a thousand users

totalRevenue
360000
activeUsers
30000
StepCalculationResult
Base ARPU360000 / 3000012.0
Annualised ARPU12.0 x 12144.0
Daily ARPU12.0 / 300.4
Revenue per thousand users360000 / (30000 / 1000)12000.0

Result: This app earns 12.0 per active user per month, or 144.0 annualised and 0.4 per day, which works out at 12000.0 of revenue for every thousand active users.

Example 3: Comparing cohorts — the annualised view

totalRevenue
240000
activeUsers
40000
StepCalculationResult
Base ARPU240000 / 400006.0
Revenue per thousand users240000 / (40000 / 1000)6000.0
Daily ARPU6.0 / 300.2
Annualised ARPU6.0 x 1272.0

Result: The later cohort contributes 6.0 per user each month, equivalent to 6000.0 per thousand users and 0.2 per day, which annualises to 72.0 per user if the month repeats.

Calculator

ARPU (per month)

$24.00

Annualised ARPU
$288.00
Revenue per thousand users
$24,000.00
ARPU per day
$0.80

Values update as you type. This calculator covers the single scenario its formula assumes — see Common Mistakes for what it leaves out.

Prefer a full-width tool? Open the ARPU calculator page.

Common Mistakes

  • Mixing revenue and user periods

    Dividing annual revenue by monthly active users, or any other mismatched pair, produces a number that means nothing. Lock the period first and confirm that both inputs describe the same window before dividing.

  • Changing the definition of active

    Counting logged-in users one month and paying subscribers the next creates a false trend that looks like a monetisation change. Keep the definition fixed, or restate historical figures when you must change it.

  • Annualising a period that is not a month

    The factor of twelve only works for monthly data. Multiplying a weekly figure by twelve understates the year badly, and multiplying an already annual figure inflates it twelvefold. Match the multiplier to the base period.

  • Using gross and net revenue inconsistently

    Refunds, discounts and credits can make gross bookings very different from recognised revenue. Pick one convention and apply it every period, otherwise a policy change will appear as a performance change.

  • Treating the average as typical

    A few high-spending customers can lift the mean well above what most users pay. Always look at a distribution or median alongside ARPU before drawing conclusions about the typical customer.

  • Confusing revenue with profit

    ARPU ignores the cost of acquiring and serving each user, so a rising figure can coexist with worsening margins. Pair it with cost per user and retention data before calling the business healthier.

FAQ

What counts as a good ARPU?

There is no universal benchmark. A mobile game, a telecoms operator and an enterprise software firm have radically different revenue models, so compare ARPU against your own history and against close competitors rather than against a single absolute target.

How is ARPU different from ARPPU?

ARPU divides revenue by all active users, while ARPPU divides by paying users only. Because the paying subset is smaller, ARPPU is always the larger number. The two together show how much of the base monetises and how much those who do pay contribute.

Why do you multiply monthly ARPU by twelve to annualise it?

Because the monthly figure is assumed to repeat for each of the twelve months in a year. It is a projection rather than a measured annual total, so seasonality or growth during the year will make the real yearly figure differ.

How does ARPU relate to customer lifetime value?

A simple lifetime value estimate multiplies ARPU by the average number of periods a customer remains active. For a 24.0 monthly ARPU and a twenty-month average lifetime, that gives roughly 480.0 of revenue per customer before costs.

Should I use the mean or the median revenue per user?

The standard ARPU formula uses the mean, which is sensitive to outliers. If a small group of heavy spenders dominates revenue, the median describes the typical user better. Reporting both is often the most honest approach.

Does ARPU include one-off payments?

It depends on whether you include them in the revenue numerator. Many teams publish a blended ARPU that includes everything and a recurring-only version that excludes one-offs, because a single large payment can otherwise distort one period.

References

  1. [1]Wikipedia, Average revenue per user — https://en.wikipedia.org/wiki/Average_revenue_per_user
  2. [2]Wikipedia, Customer lifetime value — https://en.wikipedia.org/wiki/Customer_lifetime_value
  3. [3]Wikipedia, Monthly recurring revenue — https://en.wikipedia.org/wiki/Monthly_recurring_revenue