Customer acquisition cost (CAC) is the average cost of acquiring one new customer over a given period: the marketing (and, where relevant, sales) costs divided by the number of new customers. In ecommerce and retail, CAC is most useful when it counts only first-time buyers and, ideally, only the customers that marketing actually caused, known as incremental CAC.
How to calculate customer acquisition cost
The basic formula is:
CAC = Acquisition costs ÷ Number of new customers acquired
Both the costs and the customer count must cover the same period. "New customers" should mean first-time buyers. Counting all orders or all buyers mixes acquisition with retention and makes CAC look lower than it is.
What to include in acquisition costs
Which costs to include depends on the decision the number supports. A "media-only" or paid CAC includes only advertising spend. A fully loaded CAC also includes:
- Creative and content production
- Agency fees and marketing technology
- Marketing team salaries
- Promotions and first-order discounts used to win new customers
- Sales costs, in businesses that have a sales team
Whatever the definition, state it and apply it consistently, so that CAC can be compared over time and between channels.
Calculation example
An online retailer spends 50 000 EUR on marketing in a month and acquires 1 000 first-time customers. Its blended CAC is 50 000 EUR ÷ 1 000 = 50 EUR per new customer.
Why CAC matters
- Profitability: CAC is the investment side of customer economics. A new customer is profitable only when the margin from their purchases over time exceeds the cost of acquiring them.
- Growth: in retail and ecommerce, growth depends on a steady inflow of new customers. Tracking CAC shows how expensive that growth is becoming.
- Channel and budget decisions: comparing CAC across channels and campaigns shows where the next euro buys new customers most efficiently.
CAC and customer lifetime value
CAC is most meaningful when it is compared with the value a new customer generates over time. Customer value models in marketing research treat the value of a customer base as the discounted margins from current and future customers minus the cost of acquiring them (Gupta, Lehmann & Stuart, 2004; Gupta et al., 2006).
LTV:CAC ratio: lifetime value (LTV) divided by CAC. A ratio of 3:1 or higher is a widely repeated rule of thumb. It comes from SaaS practice rather than from research, and it does not transfer directly to ecommerce, where margins, purchase frequency and the time horizon of LTV differ widely. Use the ratio with a stated LTV horizon (for example 12 or 24 months) and with gross margin rather than revenue.
Payback period: the time it takes for a new customer's gross margin to cover their CAC. In ecommerce, it is often measured as the margin from the first order plus repeat orders within a fixed window.
Types of CAC
- Blended CAC: all acquisition costs divided by all new customers. Easy to calculate, but it hides differences between channels and treats customers who would have come anyway as acquired by marketing.
- Paid CAC: paid media spend divided by new customers attributed to paid media.
- Channel or campaign CAC: spend in one channel or campaign divided by new customers credited to it. Its quality depends entirely on how customers are credited.
- Incremental CAC: spend divided by the number of new customers the channel actually caused, as measured by an experiment or a marketing mix model.
Why attributed CAC is often misleading
Platform-reported and last-click CAC divide spend by the customers a platform claims credit for. Many of those customers would have bought anyway, for example people who search for the brand name or who are already in a retargeting audience. In a large field experiment at eBay, the returns from paid search were a fraction of what non-experimental estimates suggested, and brand-keyword ads showed no measurable short-term benefit (Blake, Nosko & Tadelis, 2015). Attributed CAC for such channels therefore looks lower than the cost of a genuinely new, incremental customer.
Incremental CAC in marketing mix modeling
Marketing mix modeling (MMM) estimates the incremental sales each channel and campaign causes. When the model is built on a new-customer KPI, for example the number of first-time buyers, it gives the incremental new customers per channel, and incremental CAC follows directly:
Incremental CAC = Channel spend ÷ Incremental new customers estimated by the model
An illustrative example: a retailer spends 20 000 EUR on Meta in a month. The platform reports 500 new customers, an attributed CAC of 40 EUR. The MMM estimates that the spend caused 320 new customers, giving an incremental CAC of 62.50 EUR. The second number is the one to compare against the value of a new customer.
Splitting MMM results into new and returning customers shows which channels drive growth (new customers) and which mainly drive efficiency among existing customers. A channel can have a strong incremental ROAS (iROAS) overall but a high incremental CAC if most of its effect comes from returning customers. See how a new vs. returning customer split in MMM results helps to drive both growth and efficiency. Experiments such as geo tests and conversion lift studies can be used to calibrate these estimates; see MMM calibration.
Common CAC mistakes
- Counting all buyers as new: returning customers inflate the denominator.
- Period misalignment: spend in one period, customers in another, especially around long campaigns and seasonal peaks.
- Treating attributed customers as incremental: overstates the efficiency of channels close to the purchase.
- Ignoring customer quality: a channel with a low CAC can bring in one-time, discount-driven buyers with a low LTV.
Key takeaways
- CAC = acquisition costs ÷ new customers; define both the cost base and "new customer" explicitly.
- Attributed CAC often understates the true cost of an incremental new customer; incremental CAC from MMM or experiments corrects for this.
- Compare CAC with margin-based LTV over a stated horizon rather than relying on a fixed 3:1 rule.
Related terms
- Lifetime value (LTV)
- Predicted lifetime value (pLTV)
- Cost per conversion
- Return on advertising spend (ROAS)
- Incrementality
Further reading on Sellforte
- How a new vs. returning customer split in MMM results helps to drive both growth and efficiency
- What is causal marketing mix modeling?
- Platform ROAS vs. incremental ROAS
- MMM for ecommerce brands
References
- Blake, T., Nosko, C., & Tadelis, S. (2015). Consumer heterogeneity and paid search effectiveness: A large-scale field experiment. Econometrica, 83(1), 155–174. https://doi.org/10.3982/ECTA12423
- Gupta, S., Lehmann, D. R., & Stuart, J. A. (2004). Valuing customers. Journal of Marketing Research, 41(1), 7–18. https://business.columbia.edu/sites/default/files-efs/pubfiles/534/Valuing_Customers--JMR_2003.pdf
- Gupta, S., Hanssens, D., Hardie, B., Kahn, W., Kumar, V., Lin, N., Ravishanker, N., & Sriram, S. (2006). Modeling customer lifetime value. Journal of Service Research, 9(2), 139–155. https://doi.org/10.1177/1094670506293810
Authors
Lauri Potka is the Chief Operating Officer at Sellforte, with over 15 years of experience in Marketing Mix Modeling, marketing measurement, and media spend optimization. Before joining Sellforte, he worked as a management consultant at the Boston Consulting Group, advising some of the world’s largest advertisers on data-driven marketing optimization. Follow Lauri in LinkedIn, where he is one of the leading voices in MMM and marketing measurement.
