How to Calculate Customer Retention Rate (Service Business)

Q: How do you calculate customer retention rate?

Use ((customers at end − new customers gained) ÷ customers at start) × 100. Starting with 80 customers, ending with 92, and gaining 26 new ones gives ((92 − 26) ÷ 80) × 100 = 82.5%. Subtracting new customers is what separates retention from growth.

Q: What is the difference between retention rate and churn rate?

They are two views of the same thing. Churn rate is 100 − retention rate, the share of customers who stopped buying during the period. Retention counts who stayed; churn counts who left.

September 11, 20269 min read

Customer retention rate is the share of customers you started a period with who were still buying at the end of it: ((customers at end − new customers gained) ÷ customers at start) × 100. Churn is the mirror image, the share who left. For a service business, the number tells you how much of next year's revenue you already have versus how much you have to go win from scratch.

Most service businesses measure leads obsessively and retention not at all. That is backwards. A customer you already served costs nothing to find, already trusts your crew, and books faster than a stranger. Losing them quietly is the most expensive thing a small business does, precisely because nothing shows up on a report the day it happens.

Here is how to put a number on it.

The formula

Pick a period, a year works well for most trades, and count three things.

S — customers at the start of the period
E — customers at the end of the period
N — new customers gained during the period

Then:

Retention rate = ((E − N) ÷ S) × 100

The − N matters. Without it you are measuring growth, not retention, and a good sales month will hide the fact that your existing customers are walking out the back door.

A worked example. A landscaping company starts the year with 80 maintenance customers. It ends with 92. During the year it signed 26 new ones.

((92 − 26) ÷ 80) × 100 = 82.5%

So 66 of the original 80 stayed, and 14 left. Retention is 82.5%, churn is 17.5%. Headcount grew by 12 customers, which looks like a great year, but the business had to sign 26 new accounts to net 12. Sell 26 to keep 12, and you are running a treadmill.

Churn, and what it costs you

Churn rate is simply 100 − retention rate. In the example above, 17.5%.

Translate it into money and it stops being an abstraction. If those 14 lost customers averaged $2,400 a year each, churn cost $33,600 in annual revenue, every dollar of which had to be replaced before the business grew at all. Against a customer acquisition cost of, say, $310 per new account, replacing them also cost about $4,340 in marketing and sales effort that produced no net growth.

Two more framings worth keeping in your head:

Retention compounds. At 85% retention, a cohort of 100 customers is 72 after two years and 61 after three. At 92%, the same cohort is 85 after two years and 78 after three. Seven points of retention is the difference between rebuilding your book every few years and growing on top of it.
Retention is cheaper than acquisition. The marketing cost of keeping a customer who already knows your crew is close to zero compared to winning a new one. Every point you claw back is margin, not just revenue.

What a good retention rate looks like

It depends entirely on how often your work repeats.

Business type	Typical rebooking pattern	Retention to aim for
Recurring maintenance (landscaping, cleaning, HVAC plans)	Weekly to quarterly	85% to 95%
Repeat trade work (plumbing, electrical, handyman)	Once or twice a year	60% to 80%
Project work (roofing, concrete, remodels, fencing)	Every several years	Measure referrals instead

For project trades, a raw retention rate is close to meaningless, nobody re-roofs a house every year. Measure the repeat-and-referral rate instead: the share of this year's jobs that came from a past customer or someone they sent you. If that number is under 30%, the problem is not lead generation.

The other useful cut is retention by revenue, not by customer count. Losing ten $400 customers and losing one $9,000 customer are very different events that a headcount-based rate reports identically. Run both.

How to actually measure it

The formula is easy. Getting clean inputs is where most businesses stall, because "customer" is only obvious until you try to count them.

Define active. A customer is active if they booked work in the period. For maintenance plans, active means a live agreement. Write the definition down and use the same one every period, or your trend is noise.
Deduplicate the list. The same homeowner entered three times under three spellings looks like three retained customers, or three churned ones. One record per customer is the whole basis of the measurement.
Count at fixed dates. Same window every year, so seasonal trades compare like with like.
Tag new versus returning. You need N separated from E. If you cannot tell which jobs came from first-time customers, you cannot compute retention at all.
Review quarterly, act monthly. Churn is slow and quiet. A quarterly number tells you the trend; a monthly look at who has not booked in a while tells you who to call.

A CRM earns its keep here. When every client record carries its full job and invoice history, the counts are a filter, not a reconstruction project.

Why service customers actually leave

Almost never price. In the field trades, the recurring reasons are mundane and fixable.

You stopped showing up. For anything seasonal, the customer who is not rebooked is not loyal, they are simply unscheduled. Silence reads as "they are too busy for me."
Scheduling friction. A missed callback, a vague arrival window, a reschedule nobody communicated. Reliability beats craftsmanship in the customer's memory more often than owners expect.
A quote that never got followed up. Half of churn is a quote sitting unanswered that nobody chased. The customer did not choose a competitor, they just got tired of waiting.
Invoice friction. Confusing line items, a surprise total, an awkward payment process. Money is the last impression of the job, and it colors the whole thing.
A callback handled badly. Something going wrong is survivable. Something going wrong and being hard to reach is not.

Notice that four of the five are operations problems, not sales problems. Retention is mostly won in the boring middle of the job, not in the pitch.

The habits that move the number

Book the next visit before you leave. For any recurring work, schedule the next appointment on site. Nothing beats it, and the effort is zero.
Follow up on every open quote on a set rhythm. Day 2, day 7, day 14, then close it out. Chasing quotes is retention work, not just sales work.
Invoice the day the job ends. Fresh work is easy to pay for. A month-old invoice is an argument waiting to happen.
Watch the quiet list. Once a month, pull every customer with no booking in the last 90 days, or one full cycle for your trade. That list is your churn, before it becomes churn.
Fix callbacks fast and log what they cost. A warranty visit handled in two days buys years. One handled in three weeks ends the relationship.
Know which customers are worth keeping. Retention is not a goal for every account. A customer who pays late and burns crew hours can leave. Look at margin per customer on real hours, then decide who gets the attention.

Frequently Asked Questions

What is a customer retention rate?

Customer retention rate is the percentage of customers you had at the start of a period who were still customers at the end, excluding anyone new you gained. It measures how well you keep the business you already have, rather than how much new business you win.

How do you calculate customer retention rate?

Use ((customers at end − new customers gained) ÷ customers at start) × 100. Starting with 80 customers, ending with 92, and gaining 26 new ones gives ((92 − 26) ÷ 80) × 100 = 82.5%. Subtracting new customers is what separates retention from growth.

What is the difference between retention rate and churn rate?

They are two views of the same thing. Churn rate is 100 − retention rate, the share of customers who stopped buying during the period. Retention counts who stayed; churn counts who left.

What is a good customer retention rate for a service business?

It depends on how often your work repeats. Recurring maintenance businesses should target 85% to 95%. Trades with annual or semi-annual repeat work typically land between 60% and 80%. For project work like roofing or remodels, measure repeat-and-referral rate instead, since customers are not expected to rebook yearly.

How do I reduce customer churn in a service business?

Book the next visit before leaving the site, follow up on open quotes on a fixed schedule, invoice immediately, and review a list of customers who have not booked in a full service cycle every month. Most churn in the trades is caused by scheduling and communication gaps, not by price.

Should I measure retention by customer count or by revenue?

Both. Counting customers shows how many relationships you are holding; weighting by revenue shows what those relationships are worth. Losing one large account and losing ten small ones look identical by count and very different in the bank.

Getting started with CRMb

Retention is a records problem before it is a marketing problem. You cannot count who came back if the same customer exists three times in three spreadsheets, and you cannot tell who is worth keeping without knowing what their jobs actually cost you.

CRMb keeps every client on one record with their full job, quote, and invoice history, puts recurring visits on the crew schedule, and shows margin on your crew's real punched hours, per job and per customer, on Mac, iPad, and the web. Start a 14-day free trial and find out who is actually coming back.