The business looked healthy from every angle a board normally checks. Revenue around $300M. Positive EBITDA. A marketing function hitting its targets, reporting steady pipeline, spending roughly $3M a year on paid search to keep new customers arriving. Nothing on any dashboard was flashing red.
It was only when we asked a different question that the picture changed. Not "is the marketing working?" but "what does it actually cost to win a customer here, and what is that customer worth?" That is the unit economics question, and in most businesses it has never been answered with any rigour. This one was no exception.
When we calculated it properly, segment by segment rather than as a single blended average, one segment stood out. It was the segment almost all the paid-search budget was chasing. Its ratio of customer lifetime value to acquisition cost came in at 0.012 to 1.
To put that in plain terms: the business was paying around three thousand dollars to win a customer who would return about forty dollars over their entire life as a customer. For every dollar spent acquiring that segment, a little over one cent came back. The rest was gone.
01Why nobody had seen it
Two things kept this hidden, and both are common.
The first is the blended number. The business measured its acquisition performance across the whole customer base at once. Averaged together, the healthy customers and the value-destroying ones produced a figure that looked unremarkable. A blended ratio tells you almost nothing useful: it averages away the very thing you need to see. Underneath that comfortable average sat one segment creating real value and another quietly destroying it, and the spend was flowing overwhelmingly to the second.
The second is that nobody owned the number. Finance saw revenue and EBITDA, both fine. Marketing saw leads and pipeline, both on target. Sales saw deals closing. Each function reported coherently on its own terms, and no one was assembling those terms into the single ratio that would have shown the problem. The loop between what a customer costs to win and what they return was never closed, because closing it was not anyone's job.
A healthy blended figure can hide a segment that is destroying capital on every deal. The average is where the evidence goes to disappear.
02The belief that kept the spend flowing
There was also a story the business told itself, and it was a reasonable one. The low-value customers, the thinking went, were the future: they would grow over time into the larger, more profitable accounts. So the acquisition spend was really an investment in tomorrow's revenue.
It was a coherent hypothesis. It had simply never been tested against the data. When we looked at how cohorts actually behaved over time, the hypothesis did not survive. Below a certain size, those customers did not grow. The assumed engine of future value was, on the evidence, a cost centre with a persuasive narrative attached.
This is worth dwelling on, because it is the rule rather than the exception. The assumptions that justify acquisition spend are rarely stress-tested. They feel true, they are repeated until they harden into fact, and the cohort data that would confirm or kill them is never assembled. "These customers will grow" is a hypothesis. The cohort is the test. It usually does not survive intact.
03What changed
The fix did not require more budget. It required pointing the existing budget at the customers worth winning.
The roughly $3M a year was redirected away from the value-destroying segment and toward a clearly defined ideal customer profile, the segment the data showed was actually creating value. Within months, the reoriented approach was generating over $1M of new pipeline a month, from the same money, now spent on the right customers rather than the wrong ones.
No new capital. No heroic campaign. Just the difference between spending on conviction and spending on evidence.
The money was never the problem. The direction of the money was the problem.
04The question worth asking first
The reason this case matters is not that it is unusual. It is that the conditions that produced it are everywhere: a blended number that flatters, a metric no single function owns, and a growth story that has never met the cohort data. A business can have every dashboard green and still be destroying capital on every deal it wins in a particular segment.
The only way to know is to ask the unit economics question directly, and to calculate the answer to a standard a finance team would sign off: fully loaded acquisition cost against margin-based, discounted lifetime value, by segment, not blended. That is what the diagnostic does. It is not a marketing audit and it does not start with the creative. It starts with the number underneath everything else: is the capital you put into winning customers actually coming back?
This particular business is where the method comes from. The 0.012:1 figure is the reason it exists.
Find out whether you're asking the right question
Calibrate is a free, ten-minute, indicative read on whether your acquisition economics are likely to be a problem. It is an indication, not a measurement. Where it points to something, the Commercial Logic diagnostic measures it properly.
Note. This case is anonymised. Figures are reported as found in the engagement and are illustrative of the pattern rather than a benchmark. Lifetime value is calculated on gross margin, net of cost-to-serve and discounted; acquisition cost is fully loaded.
Related: Capital Burn Velocity, the velocity behind this 0.012:1 ratio.