LTV to CAC Ratio
Date: 2026-08-16
What a customer is worth divided by what they cost. The headline number for whether acquisition works — and it’s only as good as its two inputs, both of which are the most assumption-laden figures in the business.
What it is
The LTV:CAC ratio compares customer lifetime value against acquisition cost.
On the running model:
contribution LTV (24m) £24.00
CAC (blended) £22.50
───────
ratio 1.07 : 1
Barely above one. Each customer returns £1.07 for every £1 spent acquiring them — before any fixed costs. That’s a business acquiring at close to break-even, which is a real and common position.
The 3:1 convention
The widely-quoted benchmark, originating in SaaS, is that 3:1 is healthy, below 1:1 is unsustainable, and above 5:1 means you’re under-investing in growth.
Treat it sceptically for retail. It assumes SaaS-like economics — high gross margins, recurring revenue, long lifetimes. A retailer at 45% gross margin with a 24-month horizon is a different shape of business, and a 2:1 with fast payback can be healthier than a 4:1 realised over five years.
The “above 5:1 means under-investing” half is the more useful part: an excellent ratio can mean you’re leaving growth on the table by not bidding for customers you could profitably afford.
Why the number is so easy to inflate
Both inputs have generous definitions available, and choosing them produces very different answers:
LTV CAC ratio
revenue LTV ÷ blended CAC £80.00 £22.50 3.6 : 1 ← "healthy"
contribution LTV ÷ paid CAC £24.00 £40.00 0.6 : 1 ← losing money
Same business. Same month. One says 3.6:1, the other says you lose 40p on every customer.
Neither is dishonest in isolation, and both get quoted. The version to trust: contribution-based LTV over a bounded, observed horizon, against the CAC relevant to the decision — blended for viability, paid for channel spend — Customer Lifetime Value, Customer Acquisition Cost.
What the ratio hides
Time. It says nothing about when the money arrives. A 3:1 realised over four years and a 3:1 realised in six months are completely different businesses to fund. That’s what Payback Period adds, and it’s the constraint that actually kills companies.
Distribution. It’s a ratio of two averages over skewed populations. A high blended ratio can be one excellent segment carrying several unprofitable ones — Segmentation (analysis).
Marginal economics. It’s an average, so it says nothing about whether the next customer is worth acquiring. Marginal CAC rises with spend; average CAC doesn’t reflect that.
Incrementality. It assumes CAC bought the customers. A meaningful share would have arrived anyway, which makes true CAC lower and the ratio better — the one direction the error runs favourably — Incrementality Testing.
Using it
- Report all three numbers, never the ratio alone: LTV, CAC, and the horizon
- Segment by acquisition channel. A blended 2:1 might be 4:1 on organic and 0.8:1 on paid social, and only the segmented view is actionable — Channel Mix
- Track it as a trend, not a threshold. Improving from 1.1 to 1.4 is a real finding; being at 1.4 versus someone’s 3:1 benchmark isn’t
- Pair it with payback. The two together answer “is this worth doing” and “can we afford to do it”, and you need both
In plain terms: the ratio tells you whether acquisition makes money eventually. It doesn’t tell you whether you’ll still be solvent when eventually arrives.