A step-by-step way to work out what a new customer really costs you, fully loaded and per channel, and how long each customer takes to pay that cost back.
Customer acquisition cost (CAC), what it costs you to win one new customer, sits behind most decisions about where marketing money goes. It is also easy to get wrong, and most of the usual mistakes push it in the same direction: down. A flattering CAC lets you keep paying for channels that lose money on every customer they bring in.
Your real CAC is everything you spend on winning new customers in a period, divided by the number of new customers you won in that same period. Everything means ad spend plus agency fees, the share of staff time spent on acquisition, tools, creative and any discounts used to close the first sale. Most companies get it wrong because they count ad spend alone, divide by all customers rather than new ones, and look at one blended figure instead of one per channel.
Below you will find the formula and the mistakes that distort it, a worked example of a fully loaded CAC and the case for channel-level figures. Later sections turn CAC into a payback period, the measure that tells you whether your growth is affordable, and show which decisions change as a result. All figures are round illustrations; replace them with your own.
The formula is simple: acquisition costs in a period divided by new customers won in that period. The difficulty sits in the two inputs, and these four mistakes are the ones to check for first:
To count new customers properly, use first purchases only. In a webshop, filter for each customer's first order, which many shop platforms can flag for you. In a CRM (customer relationship management system), count accounts by the date they first became customers rather than counting closed deals, since one customer can sign several.
Decide on your definitions once and keep them. A CAC whose method changes every quarter cannot show you a trend, and the trend is often more useful than the absolute figure.
Fully loaded CAC includes every cost you would not have if you stopped trying to win new customers. That means media spend, agency or freelancer fees for acquisition work, and the share of in-house salaries spent on it. It also means tools such as landing page builders, a fair share of your CRM, creative production, and the cost of first-order discounts, free samples or free trials.
Here is an illustrative month. You spend €10,000 on ads and win 50 new customers, so your media-only CAC is €200. On top of that you pay an agency €2,500, and half of your marketing manager's time, worth €2,000 a month, goes on acquisition. Tools add €500, new ad creative costs €1,000, and a €20 first-order discount on 50 orders adds another €1,000.
Total acquisition cost is €17,000, so the fully loaded CAC is €340, seventy per cent above the media-only figure. Nothing about the campaigns changed, only the completeness of the sum.
In business-to-business (B2B) sales, where a salesperson works each enquiry, add the share of sales salaries and commission spent on winning new accounts. Keeping two figures, a marketing CAC and a total CAC including sales, shows whether a rising cost comes from the ads or from the sales process.
Use fully loaded CAC to judge whether the business model works. For comparisons between channels, use each channel's direct costs (its media spend plus the fees and creative that serve that channel alone) and leave shared overheads in the total.
Blended CAC divides total acquisition cost by all new customers, from every channel at once. It is a fair summary of the month, but it averages away the differences you need to see.
Say you spend €4,000 on Google Ads and win 50 customers, €6,800 on Meta for 20 customers and €6,000 on LinkedIn for 5, while another 9 customers arrive through referrals and organic search. Total spend is €16,800 across 84 new customers: a blended CAC of €200, which looks healthy.
Split by channel, Google Ads costs €80 per customer, Meta €340 and LinkedIn €1,200. If an average customer brings you €900 of gross profit (revenue minus the direct cost of what you sold) over the whole relationship, LinkedIn loses €300 on every customer it wins. The blend hides that completely. The nine customers who arrived without ad spend flatter the blend too, because they lower the average with no media cost behind them.
Channel figures depend on how you assign customers to channels, and no method is perfect. Use the source recorded in your CRM or shop and apply the same rule every month. Remember, too, that channels interact: someone who saw your Meta ads may later search for you on Google, and the search gets the credit.
So before cutting a channel with a high CAC, test it. Reduce its spend by a fixed amount for four to six weeks and watch the total number of new customers, not just the ones that channel claims; if the total barely moves, the cut was right.
Report blended CAC to anyone who wants the headline, and use channel CAC when deciding where next month's money goes.
Your blended CAC is a summary; your channel CAC is where the decisions are made.
CAC on its own tells you what you spent. The payback period tells you how long a customer's gross profit takes to earn that money back, and therefore how much cash your growth ties up.
For a subscription or retainer business the calculation is direct. Say your CAC is €400 and a customer pays €100 a month at an 80 per cent gross margin: each month brings in €80 of gross profit, so the payback period is five months. If customers typically cancel in month four, you lose €80 on each of them. Your return on ad spend (ROAS), the revenue the ad platforms attribute to each euro of ads, can look excellent the whole time.
For a webshop with one-off orders, measure payback in orders rather than months. Say your fully loaded CAC is €45 and an average first order of €60, excluding VAT, leaves €30 of gross profit after product cost, shipping and payment fees. The first order loses €15, so the customer only becomes profitable with a second order; your own reorder data tells you how many customers place one, and how soon.
Decide what payback your cash position can carry. A business funding growth from its own cash flow may need its money back within a few months, while one with reserves can afford to wait longer. Either way, a payback period longer than your typical customer relationship means every new customer makes you poorer.
CAC only means something next to what a customer is worth. Work out from your own records the average gross profit a new customer brings in their first twelve months. Any channel whose CAC sits above that figure needs better conversion or pricing before it gets more money.
Average CAC also hides what the next customer costs. Say Google Ads brings 50 customers at €4,000 a month, an average of €80, and raising spend to €6,000 brings 62. The average rises only to about €97, but the 12 extra customers cost €2,000 between them, roughly €167 each.
That marginal cost, not the average, should decide whether to spend more. If €167 is still well below what a customer earns you, keep going; if it is close to your limit, the channel is near its ceiling and the next euro belongs somewhere else.
Small numbers need care as well. When a channel brings fewer than about ten new customers a month, a single deal can swing its CAC wildly, so judge it on a rolling three-month figure instead.
Recalculate monthly and review the trend each quarter, comparing like with like. Note seasonal peaks next to the figures, so nobody mistakes a strong December for a better campaign.