# What is CAC and how to calculate it > The metric that decides whether your business can grow by buying customers. And the one almost everyone miscalculates because of the denominator. Source: https://www.nashmarketinglabs.com/en/blog/what-is-cac-and-how-to-calculate-it Author: Nash Marketing Labs, a paid media agency (Google Ads, Meta Ads, TikTok Ads, LinkedIn Ads). Date: 2026-08-14 --- **Short answer:** **CAC (Customer Acquisition Cost) is what it costs you to win a new paying customer. The formula is CAC = total acquisition cost ÷ new customers.** If in one month you spend €20,000 across media, agency fees and tools, and you acquire 320 first-time buyers, your CAC is €62.50. It is the metric that answers the only question that matters before you scale: does the business earn more from each customer than it costs to bring them in? And it is the one most often distorted without anyone noticing, because both sides of the division accept more than one definition. ## The formula, term by term **CAC = total acquisition cost ÷ number of new customers** **The numerator: what costs go in.** The standard definition adds up everything spent to win customers, not just the platform invoice: - **Ad spend** across all paid channels for the period. - **Agency fees or the salary of the marketing team** dedicated to acquisition. - **Tools** in the acquisition stack: analytics, feeds, email, CRM. - **Creative production**: photography, video, editing, UGC. - **Sales team**, in businesses with assisted sales where someone closes the deal. There is a narrower version, the _media CAC_ or _paid CAC_, that only counts ad spend. It is useful for judging the channel, but not for deciding whether the business is viable. What matters is not which version you pick: it is never mixing them. Comparing a media-only CAC against a target set on full-cost CAC makes you think there is margin where there isn't. **The denominator: new customers, not orders.** This is where the calculation breaks down in most accounts. An order from someone who has already bought from you does not go in the denominator. Count it and you are no longer measuring acquisition: you are measuring activity, and the number comes out artificially low. ## A worked example Ecommerce, closed month: - Google Ads and Meta Ads spend: **€17,000** - Agency fee: **€2,500** - Acquisition tools: **€500** - Total acquisition cost: **€20,000** In the back office, 500 orders. Cross-referencing against the customer history, 180 are from existing buyers. **New customers: 320.** **CAC = 20,000 ÷ 320 = €62.50** Two wrong readings of the same month: - Dividing by all 500 orders: €40. That is 36% below the real figure. - Using only the €17,000 in media over 320 customers: €53.13. Correct as a media CAC, dangerous if your target was set on total cost. With a gross margin of €55 per order, €40 says "scale"; €62.50 says "you are losing money on the first purchase and only survive if there is a repeat buy." That is literally the opposite decision, from the same data. ## What to compare CAC against CAC alone says nothing. It is always read against the value that customer leaves behind: - **Against the gross margin of the first order**, in single-purchase businesses. If CAC exceeds the order margin, every new sale makes you poorer. - **Against LTV in margin**, in businesses with repeat purchase or subscription. The standard reference is an [LTV/CAC ratio of 3 to 1](https://www.nashmarketinglabs.com/en/blog/ltv-cac-ratio): for every euro invested in acquisition, three euros of margin over the customer's lifetime. - **Against the payback period** (CAC payback). How many months it takes a customer to return what it cost to bring them in. In subscription, under 12 months is typically healthy; above 18, growth is strangling your cash flow even if the LTV/CAC ratio looks good. And it should be read **as a trend, not a snapshot**. A CAC rising month over month on the same budget is the earliest signal of channel saturation — well before ROAS starts to move. ## The four mistakes that make CAC useless - **Counting orders instead of new customers.** The most common and most expensive. It is the gap between CAC and CPA, which we cover in full in [CAC vs CPA](https://www.nashmarketinglabs.com/en/blog/cac-vs-cpa). - **Using platform conversions as the denominator.** Google Ads and Meta conversions are attributed by window and overlap across channels: adding up both platforms gives you more customers than actually exist and deflates the CAC. The denominator comes from the back office, the only source that knows who actually bought for the first time. - **Misaligning the period.** The cost is from this month; the customer may take weeks to close. In long cycles, dividing September spend by September customers mixes cause and effect. The right approach is a cohort calculation, assigning each customer to the month they entered the funnel. - **Changing the definition without documenting it.** One month with fees, another without, produces a 15% "improvement" that never happened. Fix a definition, write it down, and keep it. ## How to lower CAC (in order of real impact) Before touching a bid, the order we apply in the accounts we run: - **Fix measurement.** A high CAC almost always has broken measurement underneath: duplicate conversions, repeat customers counted as new, channels claiming the same sale. Without fixing this, everything else is optimising blind. - **Separate new customers from repeat buyers in the bid.** Google Ads lets you value new customer acquisition differently; Meta lets you exclude existing buyers. Stopping yourself from paying acquisition prices for people who were going to buy anyway is the fastest lever. - **Improve landing page conversion.** Lifting conversion from 1.5% to 2% cuts CAC by 25% without touching a single euro of spend or a single creative. - **Work the channel mix.** Average CAC hides very different channels underneath. Almost always there is one buying customers at twice the price, and the average looks fine only because another channel compensates. - **And only then, creative and bids.** Which are what everyone touches first. ## Frequently asked questions ### What does CAC stand for? CAC stands for Customer Acquisition Cost. It is the average cost of winning a new paying customer, calculated as the total acquisition cost for a period divided by the number of customers who made their first purchase in that same period. ### How do you calculate CAC? CAC = total acquisition cost ÷ new customers. The numerator includes ad spend plus agency fees, tools, creative production and the sales team if applicable. The denominator includes only customers making their first purchase — not all orders in the period. ### What is a good CAC? There is no universal good CAC: it depends on what that customer is worth. In businesses with repeat purchase or subscription, the standard reference is that LTV in margin should be at least three times the CAC. In single-purchase businesses, CAC must stay below the gross margin of the order for each new sale to contribute positively. ### Does CAC include the agency fee? In the standard definition, yes: CAC adds up the full acquisition cost, including fees, tools and creative production. The version that counts only ad spend is called media CAC or paid CAC and is useful for evaluating channels, not for deciding whether the business is viable. The critical point is to always use the same definition and compare it against targets set with that same definition. ### What is the difference between CAC and CPA? CPA is the cost of any conversion you have defined as a goal, including conversions from existing customers and events that are not sales. CAC is the cost of winning a new paying customer. In accounts with repeat purchases, CAC can be 50% or more above the CPA reported by the platform. ## More resources - [LTV to CAC ratio: what it is, how to calculate it and what's healthy](https://www.nashmarketinglabs.com/en/blog/ltv-cac-ratio) - [CAC vs CPA: the difference and which one to track](https://www.nashmarketinglabs.com/en/blog/cac-vs-cpa) --- Contact: https://www.nashmarketinglabs.com/en/request-audit - free Viability Audit, no strings attached.