Customer acquisition cost (CAC) is what you spend to win one new customer. Add up everything you spent on sales and marketing in a period, then divide by the number of new customers you won in that same period. If you spent $2,400 in a month and won 6 customers, your CAC is $400.

The formula is simple. Getting the inputs right is where most small businesses go wrong, because the biggest cost is usually the one you never write down: your own time. This guide shows the formula, two worked examples with made-up numbers, and what to do with the result.

What counts as acquisition cost

Count anything you paid, or would have paid someone, to bring in new customers during the period.

  • Ads, directory listings, lead fees and sponsorships.
  • Software used for finding buyers, sending outreach, tracking deals and answering the phone for new leads.
  • Fees paid to freelancers, agencies or sales commissions.
  • Samples, free estimates that cost you real materials or travel, and trade-show costs.
  • The share of your time, or an employee's time, spent selling.

Don't count the cost of delivering the work. Materials for a job you've already won are cost of goods, not acquisition. Also don't count marketing aimed at existing customers, like a holiday email to past clients, unless it brings in new ones.

The formula

CAC = total sales and marketing cost in a period, divided by new customers won in that period.

Use a period long enough to smooth out luck: a quarter is better than a month if your sales cycle is slow. And match the period. If you pay for ads in March and the customers close in May, a March-only calculation will look terrible and an April one will look wonderful. Neither is true.

Include your own time

This is the line people skip. If you spend 6 hours a week on estimates, follow-up and calls with prospects, that's 26 hours a month. Pick an hourly figure that reflects what an hour of your time is worth, either what you'd pay someone to do it or what you'd earn doing billable work instead. Say $40.

26 hours at $40 is $1,040 a month. It's not cash out of your account, so you can show CAC two ways: "cash CAC" without your time, and "full CAC" with it. Both are useful. Cash CAC tells you what you need in the bank. Full CAC tells you whether the whole thing is worth doing.

A worked example: a window-cleaning company

Brightline Windows, a fictional two-person company, spends the following in a month:

  • Local ads and a directory listing: $450
  • Door hangers printed and a helper to deliver them: $180
  • Outreach and follow-up software: $24.99
  • Owner's time on estimates and follow-up: 20 hours at $40 = $800

Cash spend is $654.99. With time, it's $1,454.99. Brightline wins 7 new customers that month.

Cash CAC: $654.99 / 7 = about $93.57. Full CAC: $1,454.99 / 7 = about $207.86.

Now compare to what a customer is worth. If the average new customer pays $260 for the first cleaning and returns twice more over the year at $200 each, first-year revenue is $660. Gross margin after labor and supplies is, say, 55 percent: $363. A $208 full CAC against $363 of first-year margin is a reasonable deal, with room to spare.

A second example: a small software product

Tallyfold, a fictional invoicing app at $29 a month, spends $1,500 in a month on content and a small ad test, plus $79 on outreach software, and the founder spends 30 hours on demos and replies, valued at $50 an hour ($1,500). Total with time: $3,079. They add 12 paying customers. Full CAC: $3,079 / 12 = about $256.58.

If gross margin is 85 percent, each $29 customer returns about $24.65 a month, so payback is $256.58 / $24.65, or just over 10 months.

Payback time

Payback is CAC divided by monthly gross profit per customer. It answers: how many months until this customer has paid back what it cost to get them?

For a one-time job business, payback is the first job (or the first few). For a subscription, it can take many months, which is a problem if customers leave sooner. So also work out how long customers actually stay. If Tallyfold's customers average 8 months, a 10-month payback loses money on every customer. If they average 24, it's fine. If you don't know yet, say so and watch it.

Where the number is usually wrong

A few common distortions:

  • Leaving out time, as above. A CAC with no labor cost flatters almost every small business.
  • Counting referrals as free. They're cheap, but if you gave a discount or a gift, count it.
  • Mixing periods. Costs from one quarter against customers from another.
  • Too few customers. With 3 wins, one more or fewer moves CAC by a third. Use a longer period before trusting a number.
  • Averaging channels. A blended CAC can hide that one channel costs $60 and another $900. Where you can, split by source.

You also don't need to treat any single number as truth. Industry "average CAC" figures come from companies that don't look like yours, so use your own. The number that matters is how yours changes month to month.

Turn the number into a budget

Once you trust a CAC, it sets a ceiling on what you can spend. Say Brightline wants to win 10 customers next month and has decided it will pay up to half of first-year gross profit to get a customer. First-year gross profit is $363, so the ceiling is about $181 per customer, or $1,810 for 10, time included. Its current full CAC is $208, so it's slightly over. Cash spend alone is $94 per customer, so if the owner's time is genuinely free that month, it's fine. If not, there are two levers: spend less per customer, or earn more per customer. A second cleaning booked at the first visit does the second without touching the first.

Do the same arithmetic with your own figures before you add a channel. A new channel that costs $600 a month and wins 2 customers has a CAC of $300, which loses money against $363 of margin once you count your time. The same channel winning 5 is a CAC of $120, which is excellent. You won't know which it is until you track it for a few months, so start small.

What to do with the result

Compare CAC to first-year gross profit per customer. If CAC is well below it, you can spend more to grow. If it's close, fix something before scaling: the offer, the follow-up, the channel mix. If it's above, stop and work out which input is off before spending another dollar.

The cheapest improvement is usually follow-up, because the leads are already paid for. Replying faster and following up three or four times costs time, not cash, and raises the number of customers on the bottom of the formula. An AI worker that follows up on every lead is one way to do that without adding hours. Rocketship's worker emails prospects from your own Gmail, follows up, and books the meeting, so if follow-up is your weak spot, it goes straight at the number on the bottom of the formula.

For the outreach side of the bill see how much cold outreach costs and what a sales stack costs assembled, and for a quick reference on keeping costs down, the cheapest way to do outbound. If you're weighing pay-per-lead sellers against doing it yourself, lead generation companies for small businesses compares their pricing models. And to turn a CAC into a customer target, how many customers you need to replace your salary does the arithmetic.

Knowing your CAC turns "should I spend more on this?" into a quick calculation. Rocketship finds the buyers, emails them from your own Gmail, follows up, and books the meeting on its own. Plans that get you customers start at $24.99 a month (Launch), and every monthly plan starts with a 7-day free trial, card required. See the AI workers page for what each one covers.