Subscription Metrics Formula Sheet: 25 Formulas

If you run a subscription business, you do not need more dashboards. You need a short list of metrics you calculate the same way every month, with formulas you trust. This sheet gives you 25 of them — churn, recovery, lifetime value, MRR movements, and payment health — each with the formula, what it actually tells you, and a plain worked example. Keep the definitions consistent. A metric that changes definition every quarter tells you nothing. For plain-language definitions of any term here, keep the churn glossary open alongside this sheet.

How to use this sheet

Pick a measurement window and stick to it. Monthly is the default for most of these formulas. Use the same start and end dates, the same definition of an active subscriber, and the same revenue recognition rule every time. When you change a definition, note the date you changed it and restate the prior period if you can. The examples below use round numbers and are for illustration only — they are not benchmarks. Your job is to replace them with your own numbers. If you need the step-by-step for one metric in particular, the guide on how to calculate churn rate walks through the churn formulas in depth.

Churn formulas

Churn answers one question: how much of what you had at the start of the period did you lose? Calculate it for customers and for revenue separately — they answer different questions, and the difference between them is often the most useful signal on this page. The distinction is covered in detail in revenue churn vs logo churn.

#	Metric	Formula	What it tells you
1	Logo churn rate	Customers lost in period ÷ Customers at start of period × 100	The share of accounts you lost, regardless of size.
2	Revenue churn rate	Revenue lost from churned and downgraded accounts ÷ Revenue at start of period × 100	The share of revenue you lost. If this is lower than logo churn, you are losing smaller accounts.
3	Gross churn rate	All revenue lost (churn + downgrades + contraction) ÷ Starting revenue × 100	Total leakage before expansion offsets it.
4	Net revenue churn rate	(Revenue lost − Expansion revenue from remaining accounts) ÷ Starting revenue × 100	Whether existing accounts grow enough to cover losses. Can be negative, which is good.
5	Voluntary churn rate	Customers who actively cancelled ÷ Customers at start × 100	Churn by choice — product, price, or fit problems.
6	Involuntary churn rate	Customers lost to failed payments ÷ Customers at start × 100	Churn by accident — billing problems, not product problems.
7	Involuntary share of churn	Involuntary churned customers ÷ Total churned customers × 100	How much of your churn is fixable with billing work. Recurly research puts this share at 20–40% for most subscription businesses.

Worked example, for illustration only: You start the month with 1,000 customers and lose 30. Logo churn = 30 ÷ 1,000 × 100 = 3.0%. If 10 of those 30 were lost to failed payments, involuntary churn = 10 ÷ 1,000 × 100 = 1.0%, and the involuntary share = 10 ÷ 30 × 100 = 33.3% of all churn.

Retention and lifetime formulas

#	Metric	Formula	What it tells you
8	Retention rate	100 − Logo churn rate	The mirror of churn. A 3% churn rate is a 97% retention rate.
9	Average customer lifespan	1 ÷ Monthly logo churn rate (as a decimal)	How many months a typical customer stays. At 3% monthly churn: 1 ÷ 0.03 = 33.3 months.
10	Customer lifetime value (LTV)	Average revenue per account per month × Gross margin % × Average lifespan in months	What one customer is worth over their life, after delivery costs.
11	LTV to CAC ratio	LTV ÷ Cost to acquire one customer	Whether acquisition pays back. Track the trend, not just the level.
12	Payback period	CAC ÷ (Monthly revenue per account × Gross margin %)	Months until an acquired customer covers their acquisition cost.

Worked example, for illustration only: Average revenue is $40 per month, gross margin is 80%, and monthly churn is 4%. Lifespan = 1 ÷ 0.04 = 25 months. LTV = $40 × 0.80 × 25 = $800. If CAC is $200, LTV to CAC = $800 ÷ $200 = 4.0, and payback = $200 ÷ ($40 × 0.80) = 6.25 months.

MRR movement formulas

MRR does not move for one reason. The MRR bridge breaks the movement into five pieces so you can see whether growth came from new sales, expansion, or simply losing less. Reconcile this bridge every month: starting MRR plus the five movements must equal ending MRR exactly, or one of your numbers is wrong.

#	Metric	Formula	What it tells you
13	New MRR	Sum of first-month recurring revenue from new customers	What sales and marketing added.
14	Expansion MRR	Revenue increase from existing customers (upgrades, add-ons, seats)	Whether current customers grow.
15	Contraction MRR	Revenue decrease from existing customers who stayed but paid less	Downgrades and seat reductions — an early warning.
16	Churned MRR	Revenue lost from customers who left entirely	The revenue cost of logo churn.
17	Reactivation MRR	Revenue from former customers who returned in the period	Whether winback work is producing revenue.
18	Net new MRR	New + Expansion + Reactivation − Contraction − Churned MRR	The true monthly movement. Ending MRR = Starting MRR + Net new MRR.
19	MRR growth rate	Net new MRR ÷ Starting MRR × 100	Growth as a share of the base you started with.
20	ARPU / ARPA	Total recurring revenue ÷ Active accounts	Average revenue per account. Watch it alongside churn — rising ARPU with rising logo churn means smaller accounts are leaving.

Worked example, for illustration only: Starting MRR is $50,000. New MRR is $4,000, expansion is $1,500, reactivation is $500, contraction is $800, and churned MRR is $1,700. Net new MRR = $4,000 + $1,500 + $500 − $800 − $1,700 = $3,500. Ending MRR = $53,500. Growth rate = $3,500 ÷ $50,000 × 100 = 7.0%.

Payment recovery formulas

These five formulas are the billing-health core. Track them weekly, not just monthly, because a payment problem compounds fast. Log them on a single sheet every period — the recovery KPI template is built for exactly this.

#	Metric	Formula	What it tells you
21	Payment failure rate	Failed payment attempts ÷ Total payment attempts × 100	How often charges fail at the first attempt. Count attempts, not customers.
22	Recovery rate	Failed payments eventually collected ÷ Total failed payments × 100	Of everything that failed, how much you got back. Count a payment as recovered only when the money is collected.
23	Recovered revenue	Sum of collected amounts from initially failed payments	The dollars your recovery work returned. Report alongside recovery rate — rate without dollars hides ticket-size effects.
24	Save rate (cancel deflection)	Customers who started cancelling but stayed ÷ Customers who started cancelling × 100	How often your cancel flow retains someone. Measure again at 60 days — a save that churns next month was a delay, not a save.
25	Backfill yield	Revenue collected from backlog payments ÷ Total value of backlog worked × 100	What a one-time backlog cleanup returned. Track it separately from your ongoing recovery rate — the backlog is older and colder, so the two numbers are not comparable.

Worked example, for illustration only: In a month, 2,000 payment attempts are made and 160 fail. Failure rate = 160 ÷ 2,000 × 100 = 8.0%. Of the 160 failures, 84 are eventually collected. Recovery rate = 84 ÷ 160 × 100 = 52.5%. If the average ticket is $50, recovered revenue = 84 × $50 = $4,200.

Common calculation mistakes

• Mixing windows. Dividing this month's losses by last month's starting base, or comparing a 28-day month to a 31-day month without noting it.
• Counting retries as new failures. A payment that fails, is retried, and fails again is one failed payment for recovery rate, but two failed attempts for failure rate. Decide which you are counting and label it.
• Counting a save too early. A customer who accepts an offer and cancels three days later was not saved. Use a 60-day confirmation window for save rate.
• Annualising monthly churn by multiplying by 12. Churn compounds. Annual churn = 1 − (1 − monthly churn)¹². At 3% monthly, that is about 30.6%, not 36%.
• Changing the active-customer definition mid-year. Paused, past-due, and trial accounts must be treated the same way every period, or every trend line is suspect.

The bottom line

You do not need all 25 formulas every week. You need logo churn, revenue churn, involuntary share, failure rate, recovery rate, and net new MRR calculated identically, every month, without exception. Those six numbers tell you whether you are growing, where you are leaking, and whether the leak is a product problem or a billing problem. The other 19 exist to explain the six when they move. Write your definitions down, date them, and do not quietly revise them — a boring, consistent metric beats a clever, shifting one every time.

Frequently asked questions

What is the difference between logo churn and revenue churn?

Logo churn counts customers lost as a share of customers you started with. Revenue churn counts revenue lost as a share of revenue you started with. If revenue churn is lower than logo churn, the customers leaving tend to be smaller accounts. If it is higher, you are losing larger accounts and should investigate those accounts individually.

How do I calculate customer lifetime value?

Multiply average monthly revenue per account by your gross margin percentage, then multiply by the average customer lifespan in months. Lifespan is estimated as one divided by your monthly logo churn rate expressed as a decimal. For example, a 4 percent monthly churn rate implies an average lifespan of 25 months.

What counts as a recovered payment?

A payment counts as recovered when a charge that initially failed is later collected in full. A card update by itself is not a recovery until the retried charge succeeds. Partial collections should be tracked separately so the recovery rate and recovered revenue figures stay honest.

Should I measure churn monthly or annually?

Measure monthly for operations, because monthly figures surface problems while you can still act on them. Annual figures are useful for planning and for comparing against published annual benchmarks. Convert monthly churn to annual by compounding, not by multiplying by twelve.

How often should I review these metrics?

Review payment failure rate and recovery rate weekly, because billing problems compound quickly. Review churn, MRR movements, and lifetime value monthly, using the same window and definitions each time. Review the full sheet quarterly to check that your definitions still match how your billing actually works.