Retention and acquisition economics
Would improving retention create more value than the next increment of acquisition spending?
How I use it
Measure retention that represents value
Choose the cohort entry event, return behaviour and time window. Separate continued activity, subscription retention and revenue retention. They answer different questions. Check which customer groups are staying and whether acquired customers resemble them.
Translate acquisition into lasting customers
Track spend through to a qualified, activated and retained customer. Cost per sign-up can fall while cost per valuable customer rises. Attribute costs and observation windows consistently; keep paid and organic contributions visible.
Model competing uses of money
Compare incremental contribution after variable delivery costs with acquisition spending, implementation costs and the cost of the work itself. Include delivery time and realistic adoption. For an AI product, inference and human review costs can change the result materially. Use a low, central and high scenario.
Find the assumption that changes the decision
Ask how much improvement is needed to cover the investment and what evidence would make that improvement credible. Modelled value is conditional on the change working. Extra advertising required to reach a similar audience is a comparison, not a cash saving to add to the return.
Illustrative numbers—not client results
- 50% return probability
- 625 remain
- 70% return probability
- ≈2,401 remain
Illustrative model, not client results: 10,000 initially active users, constant return probability, four consecutive weekly returns and no resurrection. 50% leaves 625; 70% leaves about 2,401.
Suppose £100,000 acquires 10,000 initially active users. In a deliberately simple model, each user must return in four consecutive weeks. A constant 50% weekly return probability leaves 625 after four returns; 70% leaves about 2,401.
The acquisition cost per user surviving all four returns is £160 versus about £41.65. This shows why retention can change acquisition economics. It assumes no resurrection and constant return probabilities; real cohort curves are rarely this simple. The example excludes the cost of achieving the improvement and is not a forecast or a measure of total WAU.
| Scenario | 50% weekly return | 70% weekly return |
|---|---|---|
| Initially active | 10,000 | 10,000 |
| After 4 consecutive returns | 625 | ≈2,401 |
| Ad cost per surviving user | £160 | ≈£41.65 |
What to watch for
- Claiming retention always matters more than acquisition.
- Using revenue as if it were contribution or cash profit.
- Counting all future improvement as caused by a project, or ignoring its delivery cost.
A brief for the next discussion.
A transparent scenario model with costs, assumptions and a break-even point.
Compare the economics
Where is your growth work getting stuck?
Use the growth diagnostic to find where to focus, or talk it through with me.