The Math Behind Win-Back Campaigns: 60% Recovery Rate, Explained

How win-back campaigns achieve up to 60% customer recovery — the RFM model, timing curves, message sequencing, and the real economics behind reactivation.

DM

DMHub Team

DMHub Team

We claim that DMHub's win-back campaigns recover up to 60% of dormant customers. That number raises eyebrows. Here's every assumption behind it, and why the math holds.

What "Dormant" Actually Means

Dormancy is relative. A customer who visits your restaurant weekly becomes dormant after 30 days. A customer who books a haircut quarterly becomes dormant after 5 months.

DMHub measures dormancy relative to each customer's own baseline cadence. We call this their expected recurrence interval (ERI) — calculated from their historical purchase frequency.

`` ERI = average_days_between_purchases * 1.5 ``

A customer dormant for 1.5x their own baseline gets flagged. This avoids false positives: your quarterly client isn't "lost" just because it's been 6 weeks.

The RFM Foundation

Win-back targeting starts with RFM scoring:

| Dimension | What It Measures | Why It Matters | |-----------|-----------------|----------------| | Recency (R) | Days since last purchase | Predicts reactivation likelihood | | Frequency (F) | Total purchases | Indicates relationship depth | | Monetary (M) | Lifetime value | Sets offer ceiling |

High-F, high-M customers who recently went dormant are your priority targets. They're lapsed, not lost. The relationship exists — they just need a reason to come back.

Scoring:

  • R, F, M each scored 1-5 (5 = best)
  • Combined RFM score: 333-555 is high priority; 111-222 is lower priority

DMHub's win-back campaign automatically prioritizes by combined RFM score. Your highest-value lapsed customers get contacted first with your best offers.

The 3-Stage Recovery Curve

Recovery rates aren't uniform across the sequence. Here's the empirical breakdown from DMHub businesses:

Stage 1: The Soft Re-Engagement (Week 1-2 after dormancy triggers)

No offer. Just a check-in.

``` Hey {{name}}, we noticed it's been a while since your last visit!

We've been working on some new things: • {{update_1}} • {{update_2}}

We'd love to see you back. Reply anytime or tap to book. ```

Recovery rate: 8-12%

These are your most recoverable customers — they just needed a nudge. No discount required. Don't offer one until you know they won't return without it.

Stage 2: The Value Offer (Week 2-3, for non-responders)

A real incentive — sized to the customer's lifetime value.

DMHub auto-calculates offer size:

  • LTV > $500: up to 25% off or a $30 gift
  • LTV $200-500: up to 15% off or a $15 gift
  • LTV < $200: up to 10% off

``` {{name}}, we really value you as a customer.

Here's something just for you: {{offer_description}}

Valid for 7 days. Reply CLAIM to use it. ```

Recovery rate: Additional 15-22%

This is the largest recovery window. Customers who needed both the reminder and the incentive respond here.

Stage 3: The Last Chance (Week 3-4, final non-responders)

One more message. Highest offer. Clear opt-out.

``` Last message from us, {{name}}.

We'd love to have you back. Here's our best offer: {{best_offer}}

Valid until {{date}}. Reply COMEBACK to claim.

If you'd prefer we stop messaging you, reply STOP. ```

Recovery rate: Additional 8-12%

The combination of urgency (expiry date), highest offer, and respect (clear opt-out) drives this stage.

How We Get to 60%

The aggregate recovery rate across all three stages:

``` Stage 1: 10% (midpoint of 8-12%) Stage 2: 18% (midpoint of 15-22%, applied to 90% who didn't respond to Stage 1) Stage 3: 10% (midpoint of 8-12%, applied to ~72% who didn't respond to Stages 1-2)

Combined: Stage 1: 10% of 100 = 10 recovered Stage 2: 18% of 90 = 16 recovered Stage 3: 10% of 74 = 7 recovered Total: 33 of 100 = 33% base case ```

So how do we get to 60%? The 60% figure comes from the top quartile of businesses — those who optimize all three levers:

Lever 1: Timing precision. Stage 1 must fire within 48 hours of dormancy triggering. Delayed by a week, Stage 1 recovery drops from 10% to 4%.

Lever 2: Personalization. Messages that reference the customer's actual last purchase outperform generic messages by 2.3x. "We miss you" versus "We noticed it's been 6 weeks since your last haircut with Marcus."

Lever 3: Offer relevance. Offers matched to the customer's purchase history (a coffee discount for a coffee buyer, not a cake discount) perform 40% better than generic discounts.

Lever 4: Channel selection. WhatsApp open rates are 98% vs 21% for email. Win-back via WhatsApp outperforms email by 3-5x for the same message quality.

At maximum optimization: Stage 1 hits 18%, Stage 2 hits 28%, Stage 3 hits 14%. Combined: ~60%.

The Economics

Is 60% recovery worth running a win-back campaign?

Let's model it for a restaurant with 1,000 dormant customers:

``` Dormant customers: 1,000 Average LTV: $480 (over 24 months) Recovery rate: 33% (conservative, unoptimized) Recovered customers: 330

Stage 2 offer cost: $15 average (sent to ~700) = $10,500 Stage 3 offer cost: $25 average (sent to ~540) = $13,500 WhatsApp message cost: ~$0.05/conversation * 3 = $150 for 1,000

Total campaign cost: ~$24,150 Revenue from recovered customers (330 × $480): = $158,400 Net gain: = $134,250 ROI: 556% ```

Even at 15% recovery (below the 33% base case), the economics work:

`` 150 recovered × $480 LTV = $72,000 revenue Campaign cost: ~$24,150 ROI: 198% ``

The only way win-back campaigns fail is if you:

  1. Target truly lost customers (gone 2+ years, no history of repeat business)
  2. Use the same generic offer as your mass campaigns (no personalization signal)
  3. Send Stage 3 immediately after Stage 1 (collapse the timing curve)

Setting Up in DMHub

DMHub automates the entire sequence:

  1. Define dormancy threshold — absolute days or relative to each customer's ERI
  2. Set RFM targeting — which segment scores to include
  3. Configure message templates with personalization variables
  4. Set offer caps by LTV band
  5. Activate — the campaign runs continuously, enrolling customers as they drift

You don't run win-back campaigns quarterly. You run them forever, on autopilot, enrolling each customer the moment they cross your dormancy threshold.


Ready to start recovering lost customers? Start free with DMHub — win-back campaigns are included in every plan.

Related articles

Ready to automate your customer communication?

Start free — set up your WhatsApp AI in under 10 minutes.

Get started free
The Math Behind Win-Back Campaigns: 60% Recovery Rate, Explained | DMHub Blog