Lifetime value gets quoted in every pitch deck and used in roughly zero operating decisions. Here's the version that actually moves the business — and the four ways it gets calculated wrong.
Lifetime value shows up in every ecommerce pitch deck and almost zero operating decisions. The founder quotes it on a board call. Marketing quotes it in CAC justification slides. Nobody on the team can tell you what changed last week to move it — because most "LTV figures" don't update faster than quarterly, and most aren't calculated correctly in the first place.
This piece is the version of LTV that actually matters for a sub-£10M / AUD $15M ecommerce business: how to calculate it honestly, four common mistakes that inflate it, and the three operating decisions it should drive.
Lifetime value (LTV, sometimes CLV — customer lifetime value) is the total gross profit a customer generates over the time they keep buying from you. Not revenue. Gross profit. Industry-shortened versions usually mean one of three things:
For most ecommerce businesses under £10M revenue, the second version — actual revenue per cohort over a defined window — is the only one that drives real decisions. The textbook formula uses too many estimates. The predictive version needs more data than you have.
If your gross margin is 55%, a customer that produces £400 of revenue produces £220 of gross profit. Comparing £400 against an £80 CAC looks like 5:1 — comparing £220 against £80 is 2.75:1, which is the real ratio. Revenue-LTV vs Cost is a vanity metric.
Even careful operators net out refunded orders. Fewer net out the cost of shipping the return, restocking it, or writing it off. Categories with 15%+ return rates (apparel especially) need this. The LTV haircut from honest return handling is usually 8–15%.
Most stores assume a 24- or 36-month "average lifespan" pulled from someone's blog post. The honest lifespan for the median DTC customer is closer to 14 months — and 60–70% of total LTV is generated in the first 6 months. Inflating the lifespan assumption inflates the LTV.
Customers from paid social have different LTV than customers from organic search. Different intent, different match quality, different cohort behaviour. A blended LTV number obscures whether your paid channels are profitable — they might look profitable in aggregate while specific cohorts are loss-making.
Illustrative DTC cohort. Same blended LTV (£226) hides 3.6× difference between channels.
For operational decisions, the LTV calculation that matters is:
This needs roughly 6 months of clean data and a willingness to wait — you don't get to know the 12-month LTV of customers acquired this week until next year. What you can do is project: cohorts from earlier months act as a leading indicator of newer ones.
If paid social customers have 6-month LTV of £180 and paid display customers have 6-month LTV of £95, and CAC is similar — every additional pound should go to paid social until the channel saturates. The blended-LTV view misses this.
Two scenarios:
| Cohort | CAC | 3-month gross profit | Payback | Action |
|---|---|---|---|---|
| Paid social, Q1 | £62 | £74 | ~2.5 months | Scale spend |
| Paid social, Q2 | £89 | £71 | ~3.8 months | Tighten targeting |
| Display, Q1 | £55 | £32 | ~5+ months | Wind down |
| Email referrals | £0 | £44 | Instant | Invest in lifecycle |
Cash-constrained businesses care about payback period more than absolute LTV. Investors care about both.
If a £20/month loyalty program generates £140 of incremental gross profit per active member per year, it's worth running. If the same program generates £40, it's not. You can't make that call without segmented LTV data — by tier, by category, by acquisition source.
You don't need an analytics platform. You need a spreadsheet and 90 minutes:
customer_id, first_purchase_month, acquisition_source (from UTM or post-purchase survey).If the cohort sizes are too small for any channel to read cleanly, expand to quarters instead of months. The shape is the same.
Three common LTV-driven decisions that are usually wrong:
3:1 is the venture-backed benchmark — for every £1 of CAC, £3 of gross profit over the lifetime. For bootstrapped businesses, 2:1 is workable if payback is under 6 months. Anything under 1.5:1 means the unit economics don't support paid acquisition at scale.
Subscription LTV is simpler — monthly recurring revenue × gross margin × average customer lifespan in months, minus support cost. The complexity is estimating lifespan; either pull it from churn data (1/monthly churn rate = average lifespan in months) or use cohort survival analysis if you have 12+ months of data.
Actual cohort LTV for operating decisions. Predicted LTV for forecasting (board reports, fundraising). They're different tools for different audiences. Predicted LTV needs at least 12 months of history to fit reliable curves; without that, predictions diverge badly from reality.
Almost always one of: (1) acquisition mix shifted to lower-quality channels, (2) first-to-second purchase rate dropped, (3) average order value fell because of more discounting. Run cohort-by-cohort. The drop will be concentrated in one bucket.
Highly category-dependent. Skincare/beauty: 12-month LTV of £140–£260 gross profit per customer. Supplements: £180–£320 (higher repeat). Apparel: £80–£160 (lower repeat). Food/beverage subscriptions: £200–£400 (frequency carries it). These are working ranges — outliers exist in both directions.
Qwrki runs analytics as part of the operating layer — the cohort LTV calculation above runs in your live dashboard, refreshed weekly, segmented by channel and product line. We don't ship LTV quarterly in a PDF — we ship the number live, the same one you'd use to decide what to spend next week.
Book a call — first hour we run a cohort export and tell you what your actual blended-vs-segmented LTV looks like.
Automated reporting is one of those phrases that sells software but doesn't change behaviour. The shift that matters is when the dashboards stop being weekly admin and start being the operating layer of the business.
Most ecommerce email is run as a campaign calendar — every Tuesday, every Thursday, every payday. Automation flips that. The brand sends because the customer did something, not because it's Tuesday.
Anyone can spend on Facebook ads. The difference between spending and running the channel comes down to four signals — and most agencies fail at least two of them.