You do not need an accounting background to run a business, but you do need to answer one question precisely: when one customer walks through the door, do you make money or lose it? That's unit economics. Everything else — growth plans, hiring, fundraising — is built on top of that answer.
Three numbers do most of the work.
CAC: what it costs to win one customer
Customer Acquisition Cost is your total sales and marketing spend over a period, divided by the customers you won in it. Spend €1,200 on ads and 10 hours of your time in a month, win 8 customers, and your CAC is somewhere north of €150 — if you count your time as free, which you shouldn't.
Two traps to avoid:
- Counting only ad spend. Your hours, the free pilots, the discounts to land the first deal — all of it is acquisition cost.
- Averaging over channels. Referrals at €10 CAC can mask the fact that your paid channel runs at €400. Know each channel's number separately.
LTV: what one customer is worth over their lifetime
Lifetime Value is the total margin — not revenue — a customer brings before they leave. For subscriptions, a serviceable approximation is:
LTV = monthly margin per customer ÷ monthly churn rate
A €30/month subscription with 70% gross margin and 5% monthly churn: €21 ÷ 0.05 = €420. Notice churn's power — cut it to 2.5% and LTV doubles to €840 without touching price or product. Early on, retention work is usually worth more than acquisition work.
Payback period: how fast your money comes back
LTV/CAC ratios get the attention, but payback period decides whether you survive long enough to collect. It's the months a customer takes to earn back their CAC.
CAC of €150 against €21/month margin = roughly 7 months underwater per customer. Now scale that: win 50 customers in a month and you've buried €7,500 that won't fully return until next year. This is how businesses with excellent unit economics die — growth consumes cash faster than customers return it. If your payback is long, your growth rate is a spending decision, not just an ambition.
Three worked examples
SaaS — €30/month tool. 70% margin (€21), CAC €150, churn 5%. LTV €420, LTV:CAC of 2.8, payback 7 months. Verdict: workable but tight — the highest-leverage moves are cutting churn and finding one cheap channel, not raising spend.
Retail — €45 average order. Margin per order €18, CAC €12, customers order 3.5 times before drifting away. LTV €63, ratio 5.3, payback immediate (first order covers it). Verdict: healthy — the game is repeat purchase rate, because a fourth order is nearly pure margin.
Services — €1,900 garden makeover. Margin per job €1,050, channel is free (estate agent referrals), but each agency relationship costs ~10 founder-hours to build and maintain. Verdict: spectacular on paper; the real unit constraint is founder time, so the number to watch is margin per delivery day, not per job.
The mistakes that actually hurt
- Using revenue where margin belongs. LTV built on revenue flatters everything by 2–4x. Always subtract the cost of delivering.
- Trusting tiny samples. Your first 10 customers — friends, early adopters, people who found you — are not representative. Treat early numbers as hypotheses.
- Ignoring your own salary. If the model only works while you're free labour, you don't have a business yet; you have a job that doesn't pay.
- Letting a blended average hide a broken channel. Unit economics are per-unit and per-channel or they're decoration.
Make them yours
Take your real price, your real costs, and your honest guess at churn, and compute all three numbers this week. They will be wrong — early numbers always are — but they'll be wrong in instructive ways, and each month of real data makes them sharper. A founder who knows their payback period to the month makes better decisions than one with a beautiful five-year projection and no idea what a customer costs.