CentriWebDatabase reactivation is a short, planned campaign to contact people already in your records, such as old enquiries, quotes that went quiet and past customers, to find the ones who now want the work. You paid to get those details once, so the job is to find who is ready. The first step is not the message. It is checking that you are allowed to send it.
What database reactivation is, and why it can pay
Most service businesses carry a list they never use. It holds people who enquired and did not book, people who got a quote and went silent, and customers from two or three years ago. Every one of them once raised a hand. Circumstances change: a budget frees up, a house sells, a deadline moves.
A reactivation campaign reaches those people with a short, relevant message and sees who answers. It differs from a win-back campaign for past customers in scope: it also covers people who never bought. It differs from ordinary follow-up because it happens once, in bulk, to a list that has been quiet for a while.
Nobody can promise a result for your list. A worked example shows the arithmetic, using invented numbers. Say you have 800 old contacts and you can lawfully message 600. If 5 per cent reply and 1 in 5 of those want a quote, that is 6 quotes. If your average job is $4,000 and you win a third of quotes, that is $8,000 from a list you already had. Change any number and the answer changes, so use your own figures.
Consent first: who you can contact
This comes before everything else, because a campaign that breaks the rules costs more than it earns.
Australia. The Spam Act 2003 has a section headed “Unsolicited commercial electronic messages must not be sent” and another that says commercial messages must contain a functional unsubscribe facility. You can read the headings in the Act on the Federal Register of Legislation. Consent is defined in the Act’s Schedule 2, so read it or take advice. Check how it applies to the email and SMS you plan to send. Calls follow separate rules: see our guide to SMS marketing laws and our guide on AI cold calling rules in Australia.
New Zealand. The Department of Internal Affairs says its spam law is meant to ensure electronic messages are sent only to customers who have consented. Commercial messages must include accurate information about the person who authorised the sending and a functioning unsubscribe facility. Details are on the DIA page. If you hold personal information from one purpose and want to use it for another, the Privacy Commissioner’s page on information privacy principle 10 sets out when that is allowed.
The United States and Canada. Different rules apply, and they are stricter in some respects. We have not covered them here. Start with the regulator’s own pages and speak to a lawyer.
This is general information and is not legal advice. Check the rules for your own case. The rules turn on facts: how you got the contact, what you told them, what they agreed to and when. Before you send:
- Sort the list by how each person came to you: a form they filled in, a call, a purchase, a lead list you bought, a business card.
- Keep a record of when and how consent was given, where you have it.
- Leave out anyone you cannot show a basis for contacting. A smaller list that is clean beats a larger one that is risky.
- Remove people who have opted out, in every system that holds them.
- Make sure every message says who you are and has a working way to opt out.
Cleaning the list before you send
Old lists are messy. Duplicates, dead emails, people who have since become customers, and people who asked you to stop. Fix these first.
- Merge duplicates so nobody gets two messages.
- Remove current customers and anyone with an open quote. They need a different message.
- Remove bounced addresses and numbers that are not mobiles.
- Tag by reason they went quiet, where you know it: price, timing, chose someone else, never replied. A short note in your CRM is enough.
- Tag by service and suburb or region, so the message can refer to something real.
If your records live across a spreadsheet, an inbox and a quoting tool, a CRM makes this much easier. Without one, a spreadsheet works for a single campaign.
Writing the message
People ignore broadcasts. They answer a short note that sounds like a person and asks a low-pressure question. Aim for three or four lines, no discount shouting, and one question that is easy to answer with a word.
Hi Sam, it’s Jo from [Business]. You asked us about a bathroom renovation in March and we never got to a firm quote. Is that still on your list for this year, or has it moved? Either answer is fine. Reply STOP to stop hearing from us.
That message names a real job, makes no offer you cannot honour, and gives an easy way out. Variations for different groups matter more than clever wording:
- Quote went quiet: ask whether the timing changed.
- Never booked: ask whether they still need it and what stopped them.
- Past customer: ask how the work is holding up and whether anything else is on the list.
Send in small batches over a few days instead of all at once. It keeps replies manageable and lets you fix a wording problem after the first fifty.
Where AI helps in a database reactivation campaign
Replies are the hard part. Send 600 messages and you may get a stream of answers: “yes”, “not now”, “call me”, “who is this”, “stop”, and the odd long story. A person can handle that for a week. After that it slips.
AI is useful here in three ways:
- Reading replies and sorting them: interested, not now, wrong person, opt out, question.
- Answering simple questions from your own information, such as service areas, typical timeframes and how to book, at any hour.
- Booking a call or visit into your real calendar, and handing the conversation to a person when a customer is ready to talk price.
Rules for the AI: it must say it is an automated assistant if asked, stop at once when someone opts out, never quote a price, and pass anything unclear or upset to a person. Opt-outs should update every list you hold, including other campaigns.
It is also useful in the preparation: drafting the three message variants, spotting that two records are the same person, and summarising an old email thread so the person who calls has context.
What a person does
- Approves the list and the message before anything goes.
- Takes the warm replies. A person who says “yes, still keen” should hear from a human the same day.
- Handles complaints and opt-outs and checks they have been honoured.
- Reviews what the AI said in the first fifty conversations, then a sample after that.
- Decides what to do with the rest: a polite last message, then leave them alone.
Not everyone on an old list wants to hear from you, and some will be irritated. That is part of why a small, careful campaign beats a large one.
Measuring it and knowing when to stop
Track four numbers: messages sent, replies, conversations that became a quote or booking, and jobs won with their value. Compare the jobs won with what the campaign cost you in time and tools. Note the opt-out count too, because a high one means the message or the list was wrong.
After the first run, tidy the list, remove those who said no, and set a date to run it again in six or twelve months. Between runs, new enquiries should get ordinary, fast follow-up so they never join the dormant pile. Our guide to how builders follow up quotes shows what that looks like for larger jobs.
If you want this built into your CRM and phone system so replies are sorted and booked for you, CentriWeb sets up automation for service businesses and looks after it afterwards. For a small list, a spreadsheet, a clear message and an hour a day of your own time will do the job.