The reliable way to get past customers booking again is to contact them with a specific, timely reason tied to the work you already did for them — a seasonal service their system needs, a maintenance interval that has come due — rather than a generic discount. They already know you and already trust you, so the barrier is not persuasion, it is that nobody has reminded them you exist. A furnace does not tell its owner to book a tune-up. Most home service businesses have never systematically contacted their own customer list, which makes it the cheapest source of booked work available to them.
Why do past customers stop calling?
Almost never because they were unhappy. Unhappy customers are memorable; the ones who quietly stop calling usually just forgot.
Think about the actual pattern. You install a water heater. The customer is delighted. Three years pass with no reason for either of you to be in touch. Then something else goes wrong, they search, and they hire whoever comes up — possibly you, possibly not, depending on what they remember.
There is no loyalty failure in that story. There is a memory failure, and memory is something you can influence.
The second reason is that most contractors are set up to acquire customers, not to keep them. The marketing budget goes to strangers. The CRM holds every person who ever paid you, and nobody has a job that involves looking at it.
What makes a past customer worth more than a new lead?
The acquisition cost was already paid, and it is not recoverable any other way.
Every customer in your database cost you something to get — ads, SEO, a referral relationship, years of building a name. That money is spent. Contacting them again costs the price of a text or an email. Contacting a stranger costs whatever a new lead currently costs in your market.
They also convert differently. A past customer does not need to be persuaded you are competent; they watched you work in their home. They do not shop three quotes for a tune-up. The conversation starts much further along.
And the timing is controllable in a way new demand is not. You cannot make someone's furnace break. You can reach every customer whose furnace you installed before the first cold snap, which is the next best thing.
What should I actually send them?
Something specific, tied to what they bought, with a reason to act now.
Reference the work you did. "We installed your water heater in 2023" proves this is not a mass blast and reminds them why they trusted you.
Give a timely reason. Seasonal service before the weather turns. A maintenance interval that has come due. An inspection that is worth doing before a known busy period.
Make replying trivial. No forms, no booking links, no app. A text they can answer with one word. Every extra step loses a share of the people who were willing.
Skip the discount. A percentage off has no deadline attached, so it does not create urgency — and it trains customers to wait for a sale. Seasonal timing does the persuading better than price does.
What not to send: a newsletter, a general "we miss you," or anything that could have been sent to any customer of any business. Generic messages get ignored precisely because they are obviously generic.
How should I segment the list?
By how long it has been and by what they bought. Those two axes cover most of what matters.
Under 18 months. Recent enough that they remember you clearly. A straightforward maintenance reminder works.
18 months to 3 years. They remember the company but not the details. Reference the specific job to re-anchor the relationship.
Over 3 years. Treat this as re-earning rather than continuing. They may have moved, replaced the system, or used a competitor. A softer message that does not assume the relationship is still live.
Then cut by service. Someone whose furnace you installed needs a different message than someone whose drain you cleared. The segment size drops and the relevance climbs, which is the trade you want.
How do I run this without wrecking my phone number?
This is where most DIY reactivation goes wrong, and the damage outlasts the campaign.
Check consent before sending. A customer who gave you their number to book a repair has not necessarily consented to marketing texts. The distinction matters, and where text consent is not documented, email is the appropriate channel.
Register for A2P 10DLC first. US carriers filter unregistered business texting. Registration takes days and is the one step whose timeline you do not control.
Send in batches, not all at once. Batching keeps volume inside carrier limits, protects your sending reputation, and means your office is not buried under three hundred replies on a Tuesday morning.
Suppress anyone who opted out. Automatically, every time.
The underlying risk is worth stating plainly: a campaign that generates complaints damages your carrier standing, and that breaks the appointment confirmations you depend on every day. Trading your operational messaging for one month's bookings is a bad deal.
What happens when the replies come in?
Faster than you expect, which is the part people underplan.
Reactivation generates conversations rather than clicks. Someone has to answer them and get people onto the schedule. If a batch goes out Monday morning and nobody looks at the inbox until Wednesday, you have annoyed a list you spent years building.
Two things to have in place before sending:
A monitored inbox with reply templates. Not one person's phone. Somewhere your office works, with the conversation history attached.
Capacity you can actually deliver. A campaign that books more work than your crews can cover is worse than no campaign. The customers you just re-engaged wait three weeks and remember that instead. Size each batch against real availability.
This is the same infrastructure that makes automated lead follow-up work, which is why the two usually get built together.
How often should I do this?
Ahead of each season rather than once.
Reactivation is not a one-time rescue of a neglected list. There is always a cohort that has drifted past the point where they would have called on their own, and that cohort refreshes continuously. Run it before each seasonal peak and the list keeps producing; run it once and you exhaust it.
The businesses that get the most from this treat it as a standing part of the calendar — the same way they treat scheduling maintenance — rather than something they reach for in a slow month. Which is exactly when it is least effective, because a slow month is usually slow for everyone and the seasonal hook is missing.
How do I measure whether it worked?
By booked jobs and their value. Not opens, not replies, not clicks.
Reply rate is a diagnostic — it tells you whether the message landed and the segmentation was right. It is not the outcome. A campaign that produces forty replies and six booked jobs is worth less than one producing fifteen replies and twelve booked jobs, and only one of those numbers pays for the crew.
Track three things per campaign:
Jobs booked, attributed back to the batch that produced them. This requires someone tagging the source when the booking happens, which is a process discipline more than a software problem.
Average value of those jobs. Reactivation often surfaces larger work than expected — a customer contacted about a tune-up who mentions the system is failing.
Opt-outs. A rising opt-out rate means the messages are landing wrong, and it is the early warning that matters most, because opt-outs are permanent.
Run those numbers against what the campaign cost and you will know within a week whether it paid, which is unusually fast feedback for anything in marketing.
Frequently Asked Questions
How big does my customer list need to be?
A few hundred past customers is enough to be worth running. What matters more than size is how complete the records are — a list with phone numbers and service history outperforms a much larger list of names and email addresses alone.
Should I offer a discount to get people back?
Usually not. A discount has no deadline, so it creates no urgency, and it teaches customers to wait for the next one. A seasonal reason to act — before the cold, before the heat — does the work better and protects your pricing.
What if customers have moved house?
Some will have. That is normal and it is why response rates are never universal. Keep the message low-pressure enough that a wrong recipient simply ignores it, and remove bounced numbers and hard opt-outs from the list as you go.
Is email or text better for reactivation?
Text gets read faster; email is on safer footing where marketing consent is not documented. Most campaigns use both — text for the consented portion of the list, email for everyone else, with an invitation to opt in to texts.
How soon will I see booked jobs?
Faster than any other channel, because the audience already knows you. Replies typically start within hours of a batch going out. Whether replies become jobs depends entirely on how quickly someone answers them.
Can I do this myself without a system?
For a small list, yes — a few hours with a spreadsheet and a phone. It stops being practical somewhere past a couple of hundred contacts, and that is the point at which the consent checking, batching and opt-out handling stop being things you can reliably do by hand.
Sitting on a customer list you have never used? Book a free strategy call and we will look at what is in it. See how database reactivation campaigns are structured.
