CAC payback is the only efficiency metric that survives a board meeting

Every SaaS board deck I have sat in has an LTV:CAC slide and every one of those slides is built on a churn number nobody has actually earned the right to assume yet. CAC payback does not have that problem. It is built from spend you already wrote and customers you already closed. That is why it survives the meeting when LTV:CAC gets waved through with a raised eyebrow and then quietly forgotten.

The exact formula, both versions

Run both every quarter, side by side. Blended payback is the number your CFO will quote in a board meeting. Paid-only payback by channel is the number that tells you which line item to cut when blended payback goes the wrong direction. Reporting only the blended figure hides a channel that has quietly doubled in cost per customer behind three others that are still healthy.

Benchmarks by segment and motion

Why LTV:CAC breaks in practice

LTV:CAC is not a bad concept. It is a bad metric to run a company on, because every input past CAC is a forecast dressed up as a fact.

1. 1

   The churn assumption is a guess wearing a decimal point

   LTV is usually calculated as average revenue per account divided by churn rate. A company two years into a segment does not have a stable churn rate, it has a handful of cohorts that have not finished proving out. Swap an assumed 2 percent monthly churn for the actual 3 percent your enterprise cohort is running and LTV drops by a third with no other input changing.

2. 2

   The discount rate is invisible in most decks

   A dollar of revenue collected in month 36 is worth less than a dollar collected today, and almost no LTV slide in a board deck discounts future cash flows at all. Skipping the discount rate systematically inflates LTV, which inflates the ratio, which makes an inefficient acquisition motion look fine.

3. 3

   Cohort immaturity hides the real number for years

   You cannot know true lifetime value of a cohort until it has actually lived out its lifetime. For a company selling annual contracts, that is a minimum of two to three renewal cycles before the number stops moving. Everything reported before that is an extrapolation from a partial curve, and the direction of the error is almost always optimistic because the customers who churn fastest are underrepresented early.

4. 4

   It is trivially easy to game with gross margin choices

   Move fully allocated support and infrastructure cost out of gross margin and into a shared overhead line, and LTV goes up with zero change to the business. Nobody audits the gross margin definition behind an LTV:CAC slide as carefully as they audit a payback number built from actual cash.

How payback interacts with pipeline coverage and sales cycle

Payback measures what happens after a customer closes. It says nothing about how long it took to get there, and that gap matters more as the sales cycle lengthens. A team with a 9 month sales cycle and a 10 month payback is looking at roughly 19 months between spending a marketing dollar and getting it back with margin. A team with a 2 month cycle and the same 10 month payback gets there in 12. Boards that only ask for payback in isolation are missing that the cash conversion cycle, not the payback number alone, is what determines how much runway a growth plan actually consumes.

Pipeline coverage compounds this. If you need 3.5x pipeline coverage to hit a bookings target and your sales cycle is nine months, the pipeline that pays back this year's plan had to exist roughly three quarters ago. A marketing team judged on this quarter's payback number while its pipeline-generation budget got cut two quarters back is being graded on a decision it did not make. Report payback alongside the pipeline coverage ratio and cycle length that produced the cohort being measured, not as a standalone score.

A worked example: two channels, same blended payback, different truth

The three levers that actually move payback

Cutting S&M spend is a fourth lever and it is the one every team reaches for first because it is the easiest to execute in a single meeting. It is also the one most likely to just move the problem two quarters downstream, because less spend now means less pipeline later, and a pipeline shortfall shows up as a payback problem again once the current cohort of deals runs out. Use the other three levers first. Cut spend only after you have confirmed the cut is coming out of demonstrably unqualified volume, not out of the pipeline that will close next quarter.

Holding marketing accountable without inviting the wrong behaviour

Grading a marketing team on blended CAC payback alone creates an obvious incentive: spend less this month, and the ratio improves immediately because the denominator effect on existing customers is unaffected while the numerator shrinks. The improvement is real on the slide and fake in the business, because it borrows against next quarter's pipeline to make this quarter's number look better.

The fix is to hold marketing to the inputs it actually controls and let payback be the outcome metric finance watches, not the target marketing is managed against week to week. Cost per qualified opportunity, opportunity-to-close rate on marketing-sourced pipeline, and pipeline coverage against the sales cycle length are all things a marketing leader can move directly. Payback is downstream of those three plus a sales execution variable marketing does not own. Measuring marketing on a number partly outside its control is how you end up with a team that starves the top of funnel to hit a target that was never fully theirs to hit.

What I would put in front of a board instead of an LTV:CAC slide

Blended CAC payback for the last four quarters, segmented by SMB, mid-market, and enterprise where volume supports it. Paid-only payback for the two largest channels, shown as a trend, not a single point. Net revenue retention from actual renewal data for the same period. Pipeline coverage and sales cycle length for the cohort that produced the payback number being reported. That set of four numbers, all built from data that already happened, tells a board more in five minutes than an LTV:CAC ratio built on an unproven churn curve tells them in twenty.

One caveat worth stating in the same breath: payback is a lagging number, and a young cohort will always flatter or punish it unfairly. A quarter where three unusually large deals closed will show a payback figure the business cannot repeat, and a quarter where two large deals slipped by three weeks will show one it does not deserve. That is why I report a four-quarter trend rather than a point, and why I refuse to change budget allocation on a single quarter's movement unless the change is larger than the historical quarter-to-quarter variance in the same series. If the variance is four months and the movement is two, nothing happened.

None of this is complicated math. It is four numbers, pulled from a CRM and a finance system that already exist, reported honestly and segmented instead of blended. The discipline is refusing to let a flattering blended average or an unprovable churn assumption stand in for the harder, more specific answer underneath it.