Acquiring a customer is hard. Keeping them is harder. Ad costs are rising and tracking is noisy. Yet many brands still over-index on acquisition while ignoring the quiet leak in the bucket: churn. This post walks through the math — in plain English — for transactional e-commerce (not subscriptions) so you can quantify the upside of improving retention.
The case for retention
Bain & Company’s classic work with Harvard Business Review showed that improving retention by even a few points can increase profits dramatically, often cited as 25–95% depending on margins and model.
Source: Bain — Zero Defections (HBR).
Retention math isn’t magic, it’s compounding. Fewer customers lost this cycle means more customers buying in the next one.
Define the key numbers
1 / Churn (if churn is roughly constant)That Expected Years approximation comes from a standard geometric-series view of retention (see: academic reference (PDF)).
Baseline example: a realistic $180 LTV
Consider a mid-size e-commerce brand with:
| Metric | Value |
|---|---|
| AOV | $60 |
| Purchase Frequency (PF) | 2 orders/year |
| Gross Margin (GM) | 50% (0.50) |
| Churn | 30% (0.30) |
Step 1 — Expected Years
If churn ≈ 30%, then Expected Years ≈ 1 / 0.30 = 3.33. To be conservative, round to 3 years.
Step 2 — LTV formula (transactional)
Step 3 — Plug the numbers
LTV ≈ $60 × 2 × 0.50 × 3 = $180
That’s your baseline LTV per retained customer under current churn.
What if you reduce churn by 5 points?
Improve churn from 30% → 25%. Then:
- Expected Years ≈ 1 / 0.25 = 4
- LTV ≈ $60 × 2 × 0.50 × 4 = $240
Result: A simple five-point retention improvement lifts LTV from $180 to $240 — a 33% increase. That extra value drops straight into profitability and supports higher CAC when needed.
Revenue impact in plain dollars
Suppose you have 10,000 active customers. A five-point improvement means 500 more customers return next year (because churn falls from 30% to 25%). If the average next-year spend is $120 (AOV $60 × PF 2), that’s:
Tie it together with Churnalysis
Retention math is powerful — but you need to know who is at risk and what to do. That’s where Churnalysis helps:
- Upload a CSV from Shopify, WooCommerce, or your CRM; no engineering required.
- See rated churn risk by customer.
- Get concise AI recommendations for win-back and re-engagement.
With clear risk signals and actions, hitting that five-point churn improvement becomes achievable, and measurable.
Sources
- Bain & Company — Zero Defections: Quality Comes to Services (Harvard Business Review): https://www.bain.com/insights/zero-defections-quality-comes-to-services-harvard-business-review-hbr/
- Retention math (geometric series / survival expectation), academic reference (PDF, link provided): https://slunik.slu.se/.../DW_2.pdf