You Got the First Sale. Where’s the Second?

The second sale shouldn’t be an accident

You finished the job. Got paid. Sent a receipt. Then waited for the customer to remember you exist.

That’s not a retention system. That’s wishful thinking with an invoice attached.

Retention means keeping customers over time. Churn means losing them. Repeat rate measures how many customers buy again within a defined window.

They’re related, but they aren’t interchangeable. A gym member cancelling is clearly churn. A homeowner who hasn’t needed another water heater isn’t necessarily a lost customer. Their first one is supposed to keep working.

Measure the relationship your business actually sells. Otherwise, you’ll invent a scary percentage and spend Monday fixing the wrong thing.

Churn works best when there’s something to cancel

For memberships, maintenance plans, and recurring service agreements, customer churn is straightforward:

Customer churn rate = starting customers lost during the period ÷ customers at the start × 100.

Say your cleaning business starts April with 200 recurring customers. During April, 20 of those customers cancel. You also sign 30 new ones.

• Starting customers: 200.
• Starting customers lost: 20.
• April churn: 20 ÷ 200 × 100 = 10%.
• April retention: 180 ÷ 200 × 100 = 90%.
• Ending customers: 210, assuming no other changes.

You grew. You also lost one in ten existing customers. Both things can be true.

Don’t subtract the new customers from cancellations. New sales can hide a retention problem, especially when your cost per lead keeps climbing.

For this calculation, retention is the percentage of your starting group still active at month-end. Define how you handle pauses and reactivations, then use that rule consistently. And label the period: 10% monthly churn is not the same thing as 10% annual churn.

For ordinary bookings, measure who actually returns

If customers book individual appointments or jobs, start with repeat rate:

Repeat rate = customers who purchased again within your chosen window ÷ customers in the original group × 100.

That original group is sometimes called a cohort. It just means customers who share a starting point, such as their first visit in January.

Imagine a service shop gets 120 first-time customers in January. Of those, 42 complete another paid visit within 90 days of their first visit.

90-day repeat rate = 42 ÷ 120 × 100 = 35%.

Count customers, not transactions. Someone returning four times still counts as one returning customer in this calculation.

Also, give everyone the full 90 days. Comparing January customers with people who first visited last Tuesday is bullshit math. The Tuesday group hasn’t had a chance.

Choose a window that fits the service. A six-week appointment cycle might justify a 90-day measure. Annual maintenance needs a longer window, with some allowance for scheduling.

For infrequent repairs, track whether customers return when another relevant need appears, renew a plan, or accept scheduled maintenance. Don’t call every quiet customer “churned.” And don’t label one minus your repeat rate as churn unless you’ve explicitly defined that inactivity as a lost relationship.

Put dollars beside the percentage

A percentage gets more useful when you attach money to it.

Using the illustrative shop example, assume the average ticket for a second visit is $180:

• Current result: 42 second visits × $180 = $7,560.
• Test target: 60 second visits × $180 = $10,800.
• Difference: 18 additional visits and $3,240 in revenue.

That target would move repeat rate from 35% to 50%. It’s a planning target, not an industry benchmark.

At an illustrative 55% gross margin, the extra revenue produces $1,782 in gross profit before any additional follow-up costs or overhead. That is not automatically net profit. Overtime, discounts, or extra administration can eat it.

Check capacity and labor percentage before promising appointments you can’t deliver profitably. Retention helps cash flow only if the extra work pays and customers actually pay you.

Your first benchmark is your own comparable customer groups, measured consistently. Then compare service types, acquisition sources, and first-visit experiences. An overall average can hide a service line where almost nobody returns.

Follow up around a reason, not a sales quota

“Just checking in” gives the customer another message to ignore.

Useful follow-up answers three questions: What’s due? Why now? How do I book?

For a maintenance business, that might be: “Your annual service is due next month. We have Tuesday morning or Thursday afternoon available. Want either one?”

Build a simple sequence:

• At checkout, explain the next recommended service and offer to book it.
• Record the service, recommended return date, contact permission, and responsible employee.
• Send a reminder shortly before that date, with one easy booking action.
• If there’s no response, send one relevant follow-up. Then stop pestering.
• After a missed expected return date, ask whether they still need the service or had a problem last time.

Respect opt-outs and applicable messaging rules. Keep clinical reminders appropriate and private.

Track completed paid bookings, not just opens or replies. Separate service complaints from timing issues. Reviews can flag problems, but direct answers often tell you what to fix. Put the sequence into your SOPs so it survives staff turnover.

Fix the missing next date first

Monday morning, pick one service customers reasonably buy again.

Pull one fully observed group of first-time customers. Calculate its repeat rate. Then check how many current customers have a recommended return date recorded.

Assign one person to own those dates, reminders, and completed-return reporting.

Start there. Before changing pricing or buying more leads, make sure customers know when to come back—and that somebody actually asks.