THE SHORT ANSWER Marketing works when a customer's lifetime value — measured in margin, from real retention — comfortably exceeds what that customer costs to acquire, computed channel by channel with every fee and hour of sales time included. Blended averages hide losing channels, and payback speed matters as much as the ratio itself.
Ask an owner if their marketing works and you'll get a feeling — "the ads seem good lately," "referrals are our best source, probably." Feelings are how budgets get wasted for years at a time. The actual answer lives in two numbers: what it costs you to acquire a customer, and what that customer is worth over their life with you. CAC and LTV. Everything else in marketing is commentary.
CAC: what a customer really costs
Customer acquisition cost is total acquisition spend divided by customers acquired. The mistakes are all in what people leave out. Full-load CAC includes ad spend, lead purchases, agency and software fees, and the sales time that turns leads into buyers. If a channel needs ten hours of your closer's week, that cost is real even though it never shows up in Ads Manager.
And it must be computed per channel. This is the one that changes businesses: blended CAC — everything averaged together — is where losing channels hide behind winning ones. Referrals at a near-zero CAC can mask a paid channel producing customers at triple what they're worth. Split it out, source by source, and the budget decisions start making themselves.
LTV: what a customer is really worth
Lifetime value is the total a customer generates over their relationship with you — and two honesty rules keep it from becoming fiction. First, margin, not revenue. A $3,000 customer with $600 of gross margin is a $600 LTV for acquisition math; comparing revenue-LTV against CAC flatters every channel you run. Second, measured retention, not hoped-for retention. Use how long customers actually stay, from your own records — not how long you'd like them to.
This is also where recurring revenue quietly changes everything. Renewals, retainers, repeat purchases — every added month of retention raises LTV without touching CAC, which means a business that keeps customers can afford to outbid everyone for acquiring them. Our own agency was built on that math: residuals meant every policy kept paying, which meant lead budgets that terrified one-and-done competitors were, for us, just arithmetic.
The ratio — and what to actually do with it
A commonly cited benchmark says LTV should run about three times CAC — treat that as a conversation starter, not a law, because the right ratio depends on your margins and your cash. The more useful lens is payback time: how many months until a customer repays what they cost? A great ratio with a two-year payback can still strangle a small business's cashflow. Know both.
Then act on the spread:
- Ratio strong, payback fast? You're underspending. Scale the winning channels before competitors find them.
- Ratio thin? Don't reflexively cut the ads — fix conversion first. Raising your close rate lowers CAC on every channel simultaneously; taking 20% to 30% is a 50% revenue lift on the same spend. It's almost always cheaper than buying more traffic.
- Ratio fine but LTV flat? Work the other lever: retention, renewals, upsells. The picture-painting upsell attached to a sale is pure LTV with zero added CAC.
The mistakes that hide losing math
- Blended CAC — losers hiding behind winners, covered above and worth repeating.
- Revenue LTV — flattering every channel by ignoring margin.
- Ignoring payback — profitable on paper, insolvent at the bank.
- No source tracking — if the CRM doesn't tag where every customer came from, none of this math is possible. Fix that first; it's an afternoon of setup.
- Set-and-forget — CAC drifts as markets and ad costs move. These are monthly numbers, not annual trivia.
Make it visible or it doesn't exist: one dashboard — CAC by channel, margin LTV, payback months — reviewed on the same monthly rhythm as the rest of your scorecard. Every consulting engagement we run installs this, because every argument about marketing dies the moment these numbers are on a wall.
A worked example, start to finish
Illustrative numbers, to make the machinery visible. A business spends $4,000 a month on a paid channel — ads, fees, and a fair share of sales time — and it produces 10 customers: CAC is $400. Each customer generates $2,500 in revenue at a 40% gross margin and, per the company's own records, a third of them buy again once: margin LTV lands around $1,330. Ratio: roughly 3.3 to 1, payback inside the first sale. That channel isn't a cost — it's a machine that turns $400 into $1,330, and the correct move is feeding it. Now the same math on their other channel: $3,000 a month producing 4 customers is a $750 CAC against the same $1,330 LTV — ratio under 2, and if the close rate on that channel's leads is weak, the first fix is the sales system, not the ad budget. Two channels, one spreadsheet, zero arguments. That's the entire point.
None of this math helps if the ratio is thin for a reason that lives outside the ad account. Before you rebuild the spreadsheet, it's worth checking which system is actually costing you the customers — a ten-question read on lead flow, speed, follow-up, close and retention takes two minutes and usually points at the same lever the CAC math does, from the other side.
Straight answers
Around 3:1 is the commonly quoted benchmark, but it's a starting point — a high-margin recurring business can thrive below it and a thin-margin one can starve above it. Payback speed matters as much as the ratio. Compute yours per channel before comparing to anyone's rule of thumb.
Tag every lead's source in the CRM, follow it through to closed customers, then divide that channel's full cost — spend, fees, and sales time — by the customers it produced over the same period. Imperfect tracking beats no tracking; start crude and tighten.
Usually not first. High CAC is more often a conversion problem wearing a marketing costume: slow speed to lead, weak follow-up, an undrilled pitch. Fix the machine the traffic lands in, and the same spend produces cheaper customers. Cut the channel only after the math still fails with a working machine behind it.
