ARPU: what average revenue per user means and how to calculate it

ARPU (average revenue per user) is the revenue a business earns per user over a period, usually a month: total revenue divided by the average number of users. It is a standard metric for telecom operators, apps, games and subscription products, because it shows whether growth comes from more users or from earning more from each one.

SaaS companies that sell to businesses usually track the same idea per account, called ARPA.

The ARPU formula

ARPU = revenue in the period / average number of users in the period

average users = (users at the start + users at the end) / 2

ARPPU = revenue in the period / average number of paying users
ARPA  = MRR / paying accounts

Use the average user count, not the end-of-period count, so a month with a big sign-up spike isn't compared against users who arrived on the last day.

A worked ARPU example

A mobile app made $42,000 in September from subscriptions and in-app purchases. It had 11,500 monthly active users on 1 September and 12,500 on 30 September, and 1,400 of them paid that month.

average users = (11,500 + 12,500) / 2 = 12,000
ARPU          = 42,000 / 12,000       = $3.50
ARPPU         = 42,000 / 1,400        = $30.00

The two numbers answer different questions. ARPU of $3.50 tells you what an average active user is worth this month, which is what you compare with the cost of acquiring a user. ARPPU of $30 tells you what a paying user spends; the ratio between them is your conversion rate at work (1,400 payers out of 12,000 users is about 11.7%).

A B2B SaaS example: $24,000 of MRR across 300 paying company accounts is an ARPA of $80. If that company adds seats-based pricing and ARPA climbs to $95 while the account count stays flat, MRR grows with no new customers.

ARPU vs ARPPU vs ARPA vs AOV

Reading a change in ARPU

A falling ARPU is not always bad news. Check the denominator before you react:

• More users, same revenue per payer: a successful free tier or a viral month brings in users who haven't paid yet. ARPU falls, ARPPU holds. Watch conversion over the next months.
• Same users, lower spend: discounts, downgrades or a cheaper plan taking share. This is the case that needs action.
• A mix shift by country or channel: users from lower-priced markets lower the average without any change in behavior. Split ARPU by segment before drawing conclusions.

Common ARPU mistakes

• Mixing free and paying users without saying so. Always state which denominator you used.
• Using the end-of-period user count. Growth months then show a falling ARPU that is only a timing effect.
• Counting annual prepayments in the month they arrive. Spread them over the months they cover, as you would for MRR.
• Including one-time revenue in a subscription ARPU. Setup fees and hardware make one month look better than the next.
• Defining "user" differently over time. If "active" changes from monthly to weekly activity, ARPU jumps without anything real happening.

Related metrics

• Monthly recurring revenue: ARPA is MRR divided by paying accounts.
• LTV to CAC ratio: lifetime value starts from ARPA times gross margin.
• Churn: if churned customers have a higher ARPU than average, revenue churn will be higher than customer churn.
• AOV: the per-order equivalent for stores.

Track ARPU automatically every week

For a subscription business on Stripe, the Stripe weekly MRR report posts MRR and the number of active subscriptions every Monday. Divide one by the other and you have revenue per subscription, close to ARPA when each customer has one subscription. For revenue per active user, weekly founder metrics puts Stripe MRR and PostHog weekly active users on the same card, so your agent can add the ratio as a third line and keep its history week after week.