Are you actually studying churn, or are you just guessing why customers leave?

I’ve been spending more time looking at churn in my business lately.

Not just “how many customers canceled.”

I wanted to know when they cancel, what stage of the relationship they’re in, what cities are weaker, what service types retain better, and where we should act before the cancellation happens.

So I used Claude to analyze recurring residential client data from the last 4 to 5 years.

• Closed jobs.
• Active jobs.
• Win-backs.
• Monthly pricing.
• Client tenure.
• Service frequency.
• Number of dogs.
• City.
• Cancellation timing.

The report ended up looking at 586 client runs from January 2022 through July 2026, including our active clients.

A few things surprised me.

The first big one: our danger zone is months 3–5. That’s where cancellation risk peaks.

Month 1 is healthier than I expected. Most new customers make it past the first month. The bigger problem happens once the newness wears off.

By month 3 or 4, some customers start re-deciding if the service is still worth it.

They’ve experienced it and paid a few invoices. The yard is no longer a disaster and life settles back down. That’s the moment where the service either becomes part of their routine, or they start thinking, “Maybe I can just do this myself.”

The data also showed that once a customer passes month 6, their expected stay jumps. That matters because it changes where we should focus retention.

A customer in month 2 needs a different touchpoint than someone who has been with us for 14 months.

Another thing that stood out: spring and summer are where we bleed the most.

April, May, and July are higher-churn months for us.

July was the worst month in the report.

That makes sense when I think about customer behavior.

People travel.
Budgets get tighter.
Kids are home.
The yard dries out.
Some customers think they’ll handle it themselves for a while.

Then a few weeks later, the yard gets away from them again. But if we already know that pattern, we can do something before it happens.

That’s the part I care about.

Churn analysis is only useful if it changes your behavior. Here’s what I’m doing with the information:

Month 1:

Focus on the first experience. Welcome touches, clean communication, day-after check-ins, making sure the customer feels confident they made the right decision.

Months 2–3:

Add a surprise-and-delight touchpoint before the cliff. This could be a free deodorizer/sanitizing treatment, a small yard report card, or a referral ask while they’re still happy.

Months 5–6:

Celebrate the 6-month mark. That milestone matters because customers who make it past month 6 are much more likely to stay long term.

Months 11–12:

Send an anniversary thank-you and possibly a prepayment offer. If someone prepays for the year, they’re carried past another risky decision window.

March:

Start the spring retention push before April and May cancellations hit. This is where we can remind customers why staying on service matters before the DIY thought gets too strong.

June:

Send vacation-related messaging before July churn hits. Remind customers that service continues while they travel, and they can come home to a clean yard.

October:

Prepare for winter budget cuts. Instead of letting someone cancel completely, we may offer a lower-frequency winter option when it makes sense.

On cancellation:

Use a cancellation survey and systematic win-back sequence. We had 58 client runs that were previous cancelled clients returning, so coming back is common enough that it deserves a process.

That’s why I think recurring service businesses need to study churn more seriously.

New leads and sales matter.

But if customers are quietly leaving at predictable points and you’re not watching for it, you’re paying to replace people you could have kept.

For me, this has been a reminder that retention can’t just be a vague goal.

Do you track when customers leave, or are you mostly finding out after the cancellation request hits your inbox?