I have watched this happen more than once, and it always has the same shape. A founder stands in front of a board, or an investor, or a room full of people who trust them, and puts up a single number: lifetime value to acquisition cost, eight to one. The room relaxes. Eight to one is wonderful; three to one is the rule, and they have nearly tripled it. Money gets raised on that slide. Hiring plans get built on it. The founder, who is not a fool, believes it, because the spreadsheet says so and the spreadsheet has never lied to them before.
Then the year happens. The customers cost what the model said, more or less, but the lifetime value never arrives. The cohorts that were supposed to pay for three years quietly leave in fourteen months. The cash that was meant to come back to fund the next wave of hiring does not come back, and payroll does not wait. Somewhere around month ten the founder is alone with the runway spreadsheet at one in the morning, recomputing, and finds that the real ratio was never eight to one. It was about one and a half to one, and it had been one and a half to one the whole time. The number on the slide had been a wish wearing the costume of a measurement.
That is the emotional centre of SaaS finance, and no one puts it on a slide: the metric you trust most is the one most willing to flatter you. So it is worth understanding exactly how a healthy-looking LTV/CAC ratio is built, and why the drivers underneath it matter far more than the ratio itself.
Where the magic number came from
The 3:1 rule has a real and respectable origin. Around 2010, David Skok of Matrix Partners published a SaaS metrics framework arguing that lifetime value should be roughly three times the cost of acquiring a customer for a recurring-revenue business to be viable, with the best companies running closer to five. He was not wrong. He was describing mature, public SaaS companies — the Salesforces and HubSpots — at steady state: stable churn, customers who genuinely stayed for years, and acquisition costs recovered well inside twelve months. Under those conditions the rule held.
The trouble is that the rule was then lifted out of those conditions and pasted onto every seed-stage startup with a login screen, almost none of which has stable churn, a proven multi-year lifetime, or a payback period anyone has actually observed. A heuristic built from companies at the end of the journey became a target for companies at the very start of it. The ratio is a floor for a mature business, not a law of nature for a young one, and treated as the latter it does real damage.
The four ways lifetime value lies
The reason the ratio flatters is almost always the same: the L is inflated. Lifetime value is the soft, forward-looking, assumption-heavy half of the fraction, and there are four standard ways it grows beyond the truth, each of them innocent on its own.
The first is using revenue instead of gross profit. Lifetime value should be built from the gross margin a customer brings, not the price they pay, because servicing them costs money. A business running 75% margins that computes LTV on revenue has overstated it by a quarter before it has done anything else wrong.
The second is the wrong churn. Founders reach for a single blended churn figure, usually the most recent and most flattering one, when early cohorts almost always churn faster than seasoned ones and revenue churn differs from logo churn. A customer base that loses 2% of revenue a month is a very different business from one that loses 2% of its logos, and averaging them into a single number hides the cohort that is bleeding.
The third is forgetting to discount. A euro of margin arriving in year five is not worth a euro today, yet undiscounted lifetime value treats it as though it were. The longer the assumed life, the larger the lie, because the value being summed sits further and further into a future that has been silently valued at par.
The fourth is the most seductive, and it is mathematical. The textbook shortcut sets lifetime value at average margin divided by churn — and as churn falls towards zero, that figure races towards infinity. Assume 1% monthly churn and you have just claimed your average customer stays for over eight years, longer than most SaaS products have existed. This is the ceiling problem: in a high-retention model, LTV can be made to look almost limitless by nudging a churn assumption a fraction of a point, and the spreadsheet will not object.
None of these requires dishonesty. A founder using revenue, a recent churn figure, no discount rate and a thin churn assumption can turn a true ratio of one and a half to one into a slide that reads eight to one, while believing every cell. That is precisely why it is dangerous. The model did not lie. It was told four small, hopeful things, and it compounded them.
The drivers that do not lie as easily
The antidote is to lean on the metrics that are harder to inflate because they are tied to cash rather than to a forecast of the distant future.
The most honest of these is the CAC payback period: how many months of gross margin it takes to earn back what you spent acquiring a customer. It is brutal because it ignores the speculative tail entirely and asks only when your money comes home. The arithmetic is unforgiving — eighteen thousand of acquisition cost against fifteen hundred of monthly revenue at 75% margin is a sixteen-month payback, which means every customer ties up real cash for well over a year before contributing a cent of profit. Multiply that by a hundred new customers a quarter and you can watch a “profitable” business run itself out of money, which is the part the LTV ratio never shows.
The second is net revenue retention — what an existing cohort is worth a year on, after churn and after expansion. Above 100%, your customers grow faster than they leave, and a base that expands is worth more than any acquisition engine; the best SaaS companies live here. The third is gross margin itself, because it sits inside every other number and decides whether growth funds itself or merely consumes capital faster. These three move slowly, resist wishful thinking, and tell you what the ratio cannot: not just whether the unit works, but whether you will still be solvent by the time it does.
Model the drivers, not the slogan
Most of this damage is structural rather than personal, and structure is the cure. FinModeler builds a SaaS financial model from the drivers themselves — acquisition cost, the real churn by cohort, margin, expansion, price — so that lifetime value is computed properly, on gross margin, with discounting, over a finite and defensible horizon, rather than conjured from a single optimistic ratio. Because the engine is deterministic, the same drivers always produce the same answer, and because you can move them, you can see what your ratio does when churn is a point worse than you hoped, or when payback stretches past the runway. The flattering number gives way to the one you can actually plan around.
So when the ratio comes out at eight to one, treat the feeling of relief as a warning rather than a result. Pull the lifetime value apart, ask which of the four hopeful assumptions is doing the work, and look at the payback period and the retention curve, which are far less willing to lie to you. The founder who survives is rarely the one with the best slide. It is the one who stopped believing the best slide in time to do something about it.
Turn your assumptions into a decision dashboard — build your SaaS model from the real drivers on FinModeler.
Have questions? Talk to us.
FAQs
What is a good LTV/CAC ratio for a SaaS company?
The widely cited benchmark is 3:1, with strong companies nearer 5:1. But it originated from mature, public SaaS at steady state, so for an early-stage startup it is a floor to grow towards rather than a target to claim, and it means little without the gross margin, churn trajectory and payback period that sit underneath it.
How should lifetime value actually be calculated?
On gross margin rather than revenue, using a realistic, cohort-aware churn rate, discounted to present value, and over a finite horizon rather than assuming customers stay forever. Each of those four corrections lowers the figure, which is the point — they pull it back towards the truth.
LTV/CAC or CAC payback — which should I prioritise?
Both, but the payback period is the one that governs survival. It tells you how long your cash is tied up before a customer turns cash-positive, which is what actually determines whether aggressive acquisition extends your runway or quietly ends it. A healthy ratio with a long payback is a cash-flow problem hiding inside a flattering number.
Sources
- Origin and intended context of the 3:1 LTV:CAC rule (David Skok, Matrix Partners, “SaaS Metrics 2.0” on the For Entrepreneurs blog, circa 2010; derived from mature public SaaS at steady state): benchmark and commentary from OpenView Partners, Bessemer Venture Partners, Meritech Capital and Marketing Case Bootcamp.
- CAC payback formula and worked example; the “ceiling problem” in high-retention LTV: OpenView (Kyle Poyar) and SaaS unit-economics references.
- Net revenue retention and LTV relationship: Bessemer Venture Partners (companies above 120% NRR running materially higher LTV:CAC).