The best marketing list you own is your job history: segments, consent and campaign math

The best marketing list you own is your job history. It knows what's installed in each house, how old it is, what you quoted and didn't sell, who let their membership lapse, and who hasn't called in two years. No list broker has that, and no ad platform can buy it. Yet most shops market with their job history about once a year, usually as one big "we miss you" email to everybody.

This post turns that history into five segments with exact definitions, matches each to the channels you're allowed to use, and runs the campaign math for a hypothetical eight-tech shop so you can see which segments actually pay. One of the popular ones doesn't, unless you change how you use it.

Five segments hiding in your job history

Here are the five segments that matter most for a residential service company, defined precisely enough that anyone in the office would pull the same list.

Twelve years for "old equipment" is a judgment call, not a standard. Pick a threshold for each system type you sell, write it down, and keep it stable so you can compare campaigns year over year.

app.tradeswork.app · Estimate follow-up board with pipeline totals

Two segments you're probably missing

"Repaired instead" losses. When an estimate for replacement is lost because the customer chose a repair, that's not a lost customer. It's a replacement deferred. Tag the reason when the estimate is lost, and those homes become the single best replacement list you'll ever have, because you've already been in the attic and they already know the system is on borrowed time.

Membership members with a failed renewal card. Some "lapsed members" never decided to leave. Their card expired. Those need a phone call and a new card, not a marketing campaign. Separate them before you mail anyone. Our membership math post covers involuntary churn.

What each channel requires before you use it

The segment says who. The channel decides what you're allowed to send. Here's the short version for U.S. contractors, with the rule behind each row.

Three practical consequences:

1. Your text list is smaller than your customer list. Only customers who opted into marketing texts belong on a promotional text. Everyone else can get mail or email. Our guide to texting customers legally covers the consent wording.
2. The 18-month clock matters for calls. A dormant customer whose last purchase was 30 months ago is no longer an "existing customer" for the do-not-call exemption. If their number is on the National Do Not Call Registry, calling them to sell is a solicitation without the exemption. Mail them instead.
3. CAN-SPAM penalties are per email. The FTC puts them at up to $53,088 for each violating message.² Use a real email tool with working unsubscribes, and keep those unsubscribes when you change systems.

Privacy: lighter than you think, but not nothing

California's privacy law applies to for-profit businesses that meet thresholds such as annual gross revenue over $26,625,000, the figure in effect since January 1, 2025.⁴ Most residential service shops are under that. That doesn't make customer data a free-for-all. Don't sell or share your list, don't upload customer lists to a vendor without knowing what they do with it, and don't use the details of someone's home for anything they'd be surprised by. CTIA's messaging guidelines tell senders not to use rented, sold or shared lists for texting at all.⁵

Clean the list before you spend a dollar

The campaign math: which segments actually pay

Here's a hypothetical eight-tech shop with about 4,200 active customers. Every rate and margin below is an assumption for illustration; your history will give you better ones after a single round.

How the costs were built: mail at $0.80 per piece including printing; email at no marginal cost; CSR calls at an assumed 6 minutes each at $25 an hour; EDDM at $0.40 with printing (USPS retail EDDM postage is $0.26 per piece⁶).

Three things jump out.

The two "known need" segments carry everything. Old equipment and unsold estimates produce about $66,000 of the $77,000 in expected margin, from 850 of the 4,530 contacts. They're the homes where you already know something is coming. Spend your first dollar and your best CSR hour there.

Lapsed members are worth more than the table shows. The $150 is first-year margin only. A restarted member brings maintenance visits that find repairs and, eventually, a replacement. If you have a lifetime value number, use it here instead; the membership revenue calculator helps.

Install neighbors barely break even as a tune-up drop. At 0.3% booking and $200 of margin, 2,400 pieces produce about $1,440 on $960 of spend. It's popular advice because it feels clever. It works better when you change the offer to replacements on streets where the houses are the same age as the one you just did, and better still when the neighbor list is filtered to your own past customers on that route rather than every door.

A year of job-history marketing on one calendar

Segments work best on a rhythm, timed to the weeks your board runs thin.

1. Every day

   Unsold estimates

   Follow-up texts or calls at day 2, 7 and 21 after presenting, to customers with job-text consent, then a handoff to a person.

2. Within 14 days of an install

   Install neighbors (filtered)

   Mail past customers on the same route first. Use a real photo of the job with the homeowner's permission.

3. 30 days after a lapse

   Lapsed members

   Card failures get a call. Deliberate lapses get a restart offer by mail and email.

4. 4 to 6 weeks before shoulder season

   Old equipment

   Mail and email replacement assessments ahead of spring and fall, when you have install capacity.

5. Twice a year

   Dormant customers

   Seasonal check offer by mail and email, plus a membership offer.

The day 2, 7 and 21 cadence for unsold estimates is the default in TradesWork's good-better-best follow-up board: unsold proposals get automatic texts until the customer replies, buys or says no, and the texts stop the moment they reply. For the shoulder-season timing, ask the business flags thin weeks a month ahead and suggests audiences such as unsold estimates, maintenance due and replacement candidates, with the rows behind each list exportable to CSV. TradesWork doesn't send bulk marketing email or print mail; use an email tool and a mail house for those, and keep the opt-outs in sync. The shoulder-season playbook covers the calendar side in more depth.

Measure each segment the honest way

Every campaign above will "work" if you count every job booked by anyone on the list. Customers on your list call you anyway. The fair measure is a holdout: randomly set aside 15% to 20% of each segment, contact the rest, and after 60 to 90 days compare booked and sold rates between the two groups. The difference is what the campaign added. Our direct mail post walks through the holdout math with an example.

1. 1

   Define the segment in fields

   Write the exact filter, including the date it was pulled.

2. 2

   Clean and split

   Run the hygiene checklist, then randomly hold out 15% to 20%.

3. 3

   Contact by allowed channels only

   Mail and email for everyone; texts only with marketing consent; calls within the 18-month window or to numbers not on the registry.

4. 4

   Tag every resulting job

   A campaign on each job so revenue rolls up by segment.

5. 5

   Compare after 60 to 90 days

   Booked rate and sold margin, contacted vs. holdout. Keep the segments that add margin; fix or drop the rest.

Do one round of this and your assumptions become your data. That's the real advantage of job-history marketing: it gets smarter every time you run it, and nobody else can copy the list.