The cost of customer acquisition formula, and the version most people get wrong
Cost of customer acquisition (CAC) is total sales and marketing spend divided by the number of new customers that spend produced, in the same period. The formula is not the hard part. The hard part is what goes into the numerator, and most calculations leave things out that should be there.
Ad spend is the obvious piece. What gets missed is everything around it: the salary or retainer of whoever runs the campaigns, the landing page and CRO work, the tools (CRM, analytics, ad platforms themselves), and any content or design time spent specifically on acquisition. Leave those out and CAC looks better than it is, which is a bad place to be making budget decisions from.
Timing is the other place the formula breaks. Spend in March against customers who signed in May understates CAC for March and overstates it for May, especially in longer sales cycles where the gap between first click and signed contract can run six to eight weeks. Attribute spend to the period it was spent in, not the period the customer happened to convert in, and use a rolling average if the sales cycle is long enough to make single-month numbers noisy.
Blended CAC hides what your paid channels actually cost
Most dashboards report one CAC number. That number is a blend of every channel, organic search, referral, outbound, paid search, paid social, and it answers a different question than the one most people think it answers. A falling blended CAC can mean your paid channels got more efficient, or it can mean organic and referral grew as a share of new customers while paid stayed exactly as expensive as before.
Channel-level CAC is the number that actually tells you whether to spend more or less on Google Ads specifically. Calculate it the same way as blended CAC, but restrict both the numerator and the denominator to customers attributed to that channel. This requires attribution that survives a multi-touch journey, which is the part most companies skip, then wonder why the paid search number looks too good or too bad to be real.
Google's own guidance on setting up conversion tracking is the right starting point if channel-level attribution is not solid yet. Without it, every CAC conversation downstream is an argument about a number nobody trusts.
What counts as a good CAC is a ratio, not a number
There is no universal good CAC. A $400 CAC is expensive for a $50 product and cheap for a $12,000 annual contract. The number that travels across businesses is the ratio between customer lifetime value (LTV) and CAC, not CAC on its own.
A commonly cited rule of thumb puts a healthy LTV to CAC ratio at 3 to 1 or higher, with payback (the time it takes CAC to be recovered in gross margin) somewhere under 12 to 18 months for most subscription businesses. Ratios much above 5 to 1 usually mean the company is under-investing in growth relative to what the unit economics could support, not that everything is going well.
Two numbers decide whether a given CAC is acceptable: gross margin, because CAC is recovered out of margin, not revenue, and retention, because a customer who churns in month four never pays back a CAC that assumed a two-year relationship. Neither shows up in the CAC formula itself, which is exactly why CAC in isolation is close to meaningless.
Where CAC actually breaks in the accounts I run
In the paid search accounts I manage, the single most common cause of a rising CAC is not the auction getting more expensive. It is the account still being measured against a conversion action, like a form fill or a lead, instead of a paying customer. A campaign can hit its cost per lead target every month and still be quietly pushing CAC up, because the leads it produces close at a lower rate than the leads a different, slightly pricier campaign produces.
In the accounts I run, fixing that gap, importing closed-deal or booked-job data back into the ad platform so bidding optimises toward revenue instead of form fills, typically moves blended CAC by a noticeably larger margin than any bid or budget adjustment does on its own. The account was never actually optimising for the thing the business cared about until that data existed.
The second recurring break is treating CAC as one number when the business is really running two or three acquisition motions at once, a self-serve funnel, an outbound sales motion, a partner channel, each with a different real cost and a different close rate. Bid and budget decisions made against a single blended figure end up starving whichever motion happens to look expensive in isolation, even when it is the one producing the best customers. Untangling that is most of what a Google Ads management engagement actually does in the first month, before any bid gets touched.
Five ways to lower CAC without just cutting quality
- Fix attribution before touching bids. A CAC number built on broken or last-click-only attribution will send budget to the wrong channel no matter how carefully you optimise inside it.
- Feed revenue back into the platform. Smart Bidding optimises toward whatever conversion action you give it. Give it a lead, and it will get very good at cheap leads that do not close.
- Cut the acquisition motion with the worst ratio, not the highest CAC. A channel with a higher CAC but a much higher LTV can be the one to fund more, not less.
- Improve close rate, not just click cost. Landing page and sales-process changes that raise close rate lower CAC without spending a dollar more on ads. Cost per lead optimisation and CAC reduction overlap more than the metrics suggest.
- Separate acquisition cost from expansion cost. Upsell and renewal spend, mixed into the same acquisition number, makes new-customer CAC look worse than it is and hides how cheap growing existing accounts actually is by comparison.
What people are actually searching around CAC right now
The table below is pulled directly from DataForSEO's US search volume data, collected on 2026-09-18. It is a useful gut check on where the confusion actually sits: the formula and the definition get searched far more than reduction tactics, which tracks with how often companies calculate CAC without a plan for what to do once they have the number.
| Search term | Monthly US volume | Difficulty (0-100) |
|---|---|---|
| cost of customer acquisition | 4,400 | 34 |
| cost of acquisition per customer | 4,400 | 25 |
| cost of customer acquisition formula | 880 | 29 |
| what is the cost of customer acquisition | 880 | 32 |
| how to calculate cost of customer acquisition | 720 | 29 |
| cost of customer acquisition definition | 210 | 30 |
| how to reduce cost of customer acquisition | 70 | 10 |
The gap between the top rows (definition and formula) and the bottom row (reduction) is roughly 60 to 1. Think with Google's measurement resources are a reasonable next stop once the formula is settled and the real question becomes what to change.
Why blended CAC moves differently than any single channel's cost
Google Ads and Meta Ads pricing move for different reasons; what Facebook ads actually cost by metric and what a Google search click costs are driven by different auctions with different competitive sets. Blended CAC absorbs both, plus organic and referral, so a company running multiple paid channels can see blended CAC hold steady while one channel gets meaningfully more expensive and another compensates. That is useful for the board deck and close to useless for deciding where to shift next quarter's budget.
If you are comparing what different providers would charge to manage that budget, how PPC pricing models actually work is worth reading before the quotes start arriving, because the management fee and the CAC it is supposed to improve are two separate conversations that get conflated constantly.
