Your Blended CAC Is Missing the Hours Your CTO Spent on a Security Questionnaire

You know your blended CAC. It is in the board deck, it informs what you are willing to spend on a lead, and it is almost certainly too low. Not because anyone fiddled it, but because it counts money that left the bank account and ignores the four hours your CTO spent last Tuesday answering someone else’s security questionnaire.

What blended CAC actually measures

Blended customer acquisition cost is total sales and marketing spend for a period divided by the customers won in it. Blended means undifferentiated: every channel, every segment, one number. Its virtue is that it is hard to game, because you cannot quietly park an inconvenient cost in a channel nobody looks at.

Its weakness is the same thing. It counts what you can see leaving the account: advertising, the sales team’s salaries, marketing, tooling, events. Salaried time spent by people who are not in the sales budget does not appear anywhere in that, because no invoice was raised and no line item moved.

The usual objection at this point is that these people are salaried and would be paid whether or not the questionnaire arrived, so the marginal cost is zero. That argument proves too much. By the same logic your sales team costs nothing, and the reason you do not believe that is that their time has an obvious alternative use. So does your CTO’s. Four hours on a procurement form is four hours not spent on the product, and the fact that the transfer never touches a ledger does not make it free.

Why this is the expensive kind of missing

Plenty of costs sit outside CAC. This one is worth chasing for three reasons that compound.

It is concentrated in your most expensive hours. A vendor questionnaire is not answered by whoever is free. Security goes to the CTO or the person acting as CISO, governance and finance to the CEO or finance director, people questions to whoever runs HR. These are the highest hourly costs in the company, and the questionnaire is specifically engineered to require them.

It scales with the deals you think are your best. Nobody sends a two-hundred-question assessment with a self-serve signup. The forms arrive on enterprise deals, which is exactly the segment whose margins you are least likely to question, because the contract value is large enough to make the answer feel obvious.

It is invisible by construction. The work is spread across five people, a week of elapsed time and several Slack threads. Nobody logs it, nobody owns it, and when it is done everyone goes back to their actual job. There is no artefact left behind that anybody would think to cost.

A cost that is spread thinly across senior people, arrives only on your largest deals, and leaves no record behind is not a rounding error. It is a blind spot with a shape.

The arithmetic, with the assumptions on the table

Take one enterprise questionnaire. Assume it needs fourteen hours of internal work: six from engineering and security, three from HR, two from finance, two from operations for the ESG section, and an hour of the CEO reading it before it goes back. At a fully loaded blended rate of eighty-five pounds an hour, that is £1,190.

Those numbers are an illustration, not a benchmark. Substitute your own and the shape holds, which is the part that matters: the cost per questionnaire is roughly fixed, because the form asks what it asks regardless of what you are charging.

That is the finding hiding in the arithmetic. A fixed cost per deal is regressive. It is a rounding error on your flagship contract and a real bite out of the mid-sized ones, and because it is invisible you have been ranking those two on numbers that do not include it.

The larger cost is usually the calendar

The hours are the part you can count. The elapsed time is often worth more.

A questionnaire that sits in an inbox for a week does not add hours to anybody’s timesheet, but it adds a week to the sales cycle. A longer cycle means fewer deals closed per period from the same pipeline, and CAC has the number of customers in its denominator. Slower closing raises the cost of every acquisition without anyone working an extra minute.

This is why the fix is rarely “spend fewer hours”. It is usually to stop the handovers, which is a different problem with a different solution.

How to count it without timesheet theatre

The obvious objection to measuring internal time is that it turns into surveillance, people start logging fiction, and the data becomes worse than nothing. That is a real failure mode and it comes from measuring the wrong unit.

You do not need an account of anybody’s day. You need the duration of work on one questionnaire. Three things are enough:

Note what is absent. No task codes, no per-person productivity, no approval workflow. The question is what this deal cost to win, and answering it does not require knowing how any individual spent their afternoon.

What the number changes once you have it

Costing questionnaire time is only worth doing if it changes a decision. It changes four.

Which segment you chase. If the mid-market deals that looked marginally profitable were carrying an uncounted four-figure cost each, some of them were not profitable. That is a targeting decision you have been making without the input.

What a reusable answer is worth. The second questionnaire should be far cheaper than the first, because most of it has been answered before. Without a cost per questionnaire you cannot show that curve bending, so investment in an answer bank stays a matter of opinion.

Whether to bid at all. Some questionnaires arrive on deals that were never going to close. Knowing the median cost of answering one gives you a real number to weigh against the probability, instead of the current method, which is that somebody sighs and does it anyway.

Where the time actually goes. Hours attached to a questionnaire show which sections are expensive. If the ESG section costs three times what the security section does, that is where the next reusable answer should come from.

Where to start

Turn it on for one quarter and change nothing else. You need a handful of completed questionnaires with real hours against them before the number means anything, and the first one will be your most expensive because nothing has been answered before.

Then put the cost per questionnaire next to your blended CAC and see whether the gap is large enough to matter. If it is not, you have spent a quarter of near-zero effort and can stop. If it is, you have found a cost that was shaping your pricing and your targeting from somewhere outside the number you were looking at.