How to Calculate and Track Customer Churn for a Service Business

05 September 2026·5 min read

Quick answer: Customer churn rate is the percentage of customers you lost over a set period, calculated as (customers lost during the period ÷ customers you had at the start of the period) × 100. You should also track dollar or revenue churn, which measures lost recurring revenue rather than headcount, since one lost customer can matter a lot more or less than another depending on what they were paying. There's no single "good" churn number that applies to every business — it depends heavily on industry, price point and contract length — but tracking it consistently, monthly, and split by cause is what actually makes it useful. 📈

Churn is one of those numbers businesses either ignore completely or calculate once a year and call it done. Neither gets you anywhere. I think churn deserves the same monthly discipline as your bank balance, because a small leak compounds fast — losing five percent of customers a month feels tiny until you realise that's over half your base gone within a year if nothing changes. I genuinely love how much clarity this one number gives a business once it's tracked properly 💖 — it tells you whether your marketing is filling a bucket with a hole in the bottom, and roughly how big that hole is.

What most businesses get wrong

Most businesses calculate churn annually, or never, so problems get spotted months after they actually started. Plenty only count outright cancellations and ignore downgrades — which is really revenue churn hiding inside a headcount number that looks perfectly fine. A lot lump voluntary churn (a customer chose to leave) in with involuntary churn (a payment simply failed), as if both need the same fix — they don't, and treating them the same wastes effort on the wrong problem. Many compare their churn rate to an industry benchmark pulled from a random source without checking it's even remotely the same business model. And most never segment by cohort, so when churn spikes they can't tell whether it was a pricing change, an onboarding tweak, or something else entirely.

Churn calculation worksheet

Customer churn rate = (Customers lost in period ÷ Customers at start of period) × 100

Revenue churn rate = (Recurring revenue lost in period ÷ Recurring revenue at start of period) × 100

Net revenue churn = (Recurring revenue lost − Expansion revenue from existing customers) ÷ Recurring revenue at start of period × 100 — this one can go negative, which is a genuinely good sign

Monthly tracking template — columns to keep in a simple spreadsheet:

Month
Customers at start / New customers gained / Customers lost / Customers at end
Customer churn rate %
Recurring revenue at start / Recurring revenue lost
Revenue churn rate %
Voluntary vs involuntary split (%)
Note / likely cause (price rise, service issue, competitor, payment failure)

Honest nuance: there is no universal "good" churn benchmark. A monthly subscription business, a high-touch annual-contract B2B firm, and a low-cost app all compare on completely different scales — your own trend over time is the most useful benchmark you have.

A subscription-style service business with monthly retainers: switched from annual to monthly churn tracking and noticed a spike concentrated in customers who'd joined via one specific promotion. It traced back to a mismatch between what the promo promised and what the standard service delivered — an onboarding messaging fix for that segment specifically resolved most of it.

A membership-based local business with a seasonal dip: saw churn rise every year in the same quiet months and had been treating it as a mystery. Once tracked by cohort, it was clearly seasonal — people pausing over holidays — rather than a quality problem, so the fix was a pause option instead of a retention campaign aimed at the wrong cause entirely.

A B2B firm with annual contracts and a long sales cycle: annual-only churn tracking meant a real problem — a competitor undercutting at renewal time — wasn't visible until contracts came up, months after it started. Moving to a rolling quarterly renewal-risk review, flagging accounts showing early disengagement, caught the next round of at-risk renewals far earlier.

Setting this up without expensive software

Pick your period — monthly is the minimum useful cadence for most service businesses
Build the spreadsheet template above, or pull the same columns from your CRM or billing platform if it already tracks recurring customers
Split churn into voluntary and involuntary — the fixes are completely different, one's a retention conversation, the other's a billing or dunning fix
Track by cohort (the month or quarter someone joined) where you can, not just as one blended monthly number
Review monthly, and treat any month where churn jumps noticeably as worth a quick "why" investigation while it's fresh
Revisit your formula assumptions — are you counting downgrades, partial-month cancellations — at least once a year so your number stays comparable over time

💡 Watch involuntary churn separately — it's often the easiest win. A chunk of "lost" customers in most subscription-style businesses didn't choose to leave at all; their payment simply failed and nobody followed up. A basic retry-and-reminder process recovers some of these before you ever need a retention conversation.

Mistakes to avoid

Calculating churn once a year instead of monthly
Blending voluntary and involuntary churn into one number
Ignoring downgrades because "they're still a customer"
Comparing your number to a benchmark from an unrelated business model
Not tracking by cohort, so a spike can't be traced to its actual cause
Treating churn as a lagging report instead of a monthly working number

Frequently asked questions

What's a "good" churn rate?

Honestly, there isn't one universal good number — it varies hugely by industry, contract length and price point, so the most useful comparison is your own business's trend over time, not an external benchmark.

Should I count downgrades as churn?

They're not customer churn (the customer's still there), but they are revenue churn — track both so a wave of downgrades doesn't hide inside a headcount number that still looks healthy.

How often should I actually check this?

Monthly at minimum for most service businesses. Annual checks let problems run for months before anyone notices them.

Does reducing churn matter more than gaining new customers?

Both matter, but reducing churn is often cheaper to fix than it is to keep replacing lost customers with new acquisition spend — though a very early-stage business with almost no customer base yet may reasonably prioritise acquisition first.