Customer reactivation model: revenue and cost per job from past customers
By Moe Ameen, founder of BILT · Published · Updated
This reactivation model estimates jobs from a win-back campaign as past customers x reactivation rate, and cost per job as campaign cost / jobs. With illustrative inputs (2,000 past customers, a 6% response, $500 plus $0.30 per contact), it gives 120 jobs at about $9 each. The comparison to new-lead cost is only as good as the reactivation rate you plug in.
Key takeaways
- Jobs = past customers x reactivation rate. Revenue = jobs x average ticket.
- Cost per job = (fixed cost + per-contact cost x list size) / jobs.
- The 3%, 6% and 9% reactivation rates, the $400 ticket and the costs are illustrative.
- HBR (2014) reports new customers cost 5 to 25 times more to acquire than keeping existing ones, depending on the study.
- Measure your own rate with a test send to a slice of the list before you trust the model.
The previous version of this page treated a 3% to 9% reactivation rate as 'realistic' and a $150 to $300 new-job cost as a home services norm. Neither had a source. The arithmetic was fine; the inputs were guesses presented as facts.
This version keeps the arithmetic and labels every input. It also fixes the campaign cost so it grows with list size, since texting 8,000 people costs more than texting 1,000. For your own numbers, the marketing ROI calculator at /tools/marketing-roi-calculator handles the revenue side.
How does the model work?
Formulas: jobs = past customers x reactivation rate; revenue = jobs x average ticket; campaign cost = $500 fixed + $0.30 per contact; cost per job = campaign cost / jobs; new-lead comparison = jobs x $200.
Assumption 1, reactivation rates of 3%, 6% and 9%. Illustrative. We found no public, measured reactivation rate for home services. Run a test send to a few hundred past customers and count booked jobs.
Assumption 2, a $400 average ticket. Illustrative. Use your own average invoice for the service you're promoting.
Assumption 3, $500 fixed plus $0.30 per contact. Illustrative, standing in for setup time plus a few messages per person. Check your actual per-message rates; for example, the BILT CRM pricing page lists per-segment SMS rates by plan.
Assumption 4, $200 to acquire a new job from paid leads. Illustrative. Use your own cost per booked job from the cost per lead calculator at /tools/cost-per-lead-calculator.
Context: Harvard Business Review (Amy Gallo, October 2014) reports that acquiring a new customer is 'anywhere from five to 25 times more expensive than retaining an existing one,' depending on the study and industry.
What does the model show?
| Scenario | Jobs booked | Revenue | Campaign cost | Cost per job | Same jobs at $200 each |
|---|---|---|---|---|---|
| 1,000 customers at 3% | 30 | $12,000 | $800 | $26.67 | $6,000 |
| 1,000 customers at 6% | 60 | $24,000 | $800 | $13.33 | $12,000 |
| 2,000 customers at 6% | 120 | $48,000 | $1,100 | $9.17 | $24,000 |
| 2,000 customers at 9% | 180 | $72,000 | $1,100 | $6.11 | $36,000 |
| 4,000 customers at 6% | 240 | $96,000 | $1,700 | $7.08 | $48,000 |
| 8,000 customers at 6% | 480 | $192,000 | $2,900 | $6.04 | $96,000 |
Jobs = past customers x rate. Revenue at $400 per job. Campaign cost = $500 + $0.30 per contact. New-lead comparison at $200 per job. All inputs illustrative.
What does a reactivation campaign cost per job?
In the model, 2,000 past customers at 6% gives 120 jobs for $1,100, or $9.17 per job. At 3% on a 1,000-customer list it's $26.67 per job. Cost per job falls fast as the response rate rises.
The comparison column shows the same jobs bought at the assumed $200 each: $24,000 for 120 jobs. That gap is large, but it rests on two assumptions you have to check: your real response rate and your real new-job cost.
Why does the gap exist at all?
Past customers already know you, and you already have their contact details, so you aren't paying for ads, clicks or leads. That's the logic behind the HBR finding that retention is cheaper than acquisition.
The HBR range (5 to 25 times) is wide because it comes from different studies and industries. It supports the direction of this model, not any specific cost per job.
How does list size change the result?
Revenue grows in line with the list: 8,000 customers at 6% gives 480 jobs and $192,000 in the model. Cost per job drops toward the per-contact cost divided by the rate ($0.30 / 0.06 = $5) as the fixed cost spreads out.
The catch is list quality. Old records go stale, people move and numbers change. A large, old list can respond at a much lower rate than a recent one.
What can't this model tell you?
It doesn't measure incrementality. Some past customers would have called you anyway, so a campaign's booked jobs overstate the extra work it created. Hold back a random slice of the list and compare.
It also counts revenue, not profit, and ignores capacity: 480 extra jobs in a month may be more than your crews can do. Texting past customers is covered by TCPA rules (47 CFR 64.1200 treats a text as a call), so check consent before you send. BILT's home services version is on a waitlist at /for/home-services; it isn't live yet.
Who is this for, and who should skip it?
Best for
- Home services businesses with a few years of customer records and phone or email consent
- Deciding whether a win-back test is worth an afternoon of setup
- Seasonal trades (HVAC, irrigation, gutters) where past customers need the service again
Not ideal for
- New businesses with a small customer list. There's nothing to reactivate yet.
- One-time jobs customers rarely repeat, where the reactivation rate will be low
- Lists without consent to text. Fix consent first or use mail and email instead.
What else do people ask?
What reactivation rate should I expect?
We found no public, measured rate for home services. Send to a few hundred past customers and count booked jobs; that's your rate.
Is keeping customers really cheaper than finding new ones?
HBR (2014) reports acquisition costs 5 to 25 times more than retention, depending on the study and industry. Your own cost per new job is the number to compare against.
How do I know the campaign created the jobs?
Hold out a random slice of the list, send to the rest, and compare booking rates over the same period. The difference is the campaign's real effect.
Can I text my past customers?
Under 47 CFR 64.1200 a text is treated as a call, so TCPA consent, calling-hour and opt-out rules apply. Check what consent you collected when they booked.
The takeaway
Reactivation can be the cheapest work a home services shop books, because the customer and the contact details are already yours. How cheap depends on a response rate no public dataset gives you. Test a slice of your list, hold out a control group, and put your real numbers into the model.
How do we know this?
We rebuilt this page as a model on November 4, 2026, changed the campaign cost to scale with list size, recomputed every cell with a script, and read the HBR retention article, the TCPA rule text and the BILT CRM pricing page. No BILT customer data was used.
Sources
- Harvard Business Review: The Value of Keeping the Right Customers (Amy Gallo, October 2014) (accessed November 4, 2026)
- 47 CFR 64.1200 (TCPA rules), via Cornell LII (accessed November 4, 2026)
- BILT CRM pricing page (accessed November 4, 2026)