Customer Acquisition Cost: Calculate Ecommerce CAC

+

Your ad dashboard says a purchase cost $20. Your finance sheet says a new customer cost $45. Before choosing which number is wrong, check what each one counts. The ad figure may cover one platform's reported purchases; the business figure can include more acquisition costs and only first-time customers.

Customer acquisition cost, or CAC, is acquisition spending divided by the number of new customers gained in the same period. Define the costs and the customers before you calculate it. Then compare the result with an ad-platform cost per purchase only after you have checked the scope of each number.

One number, three decisions about scope

Start with a time period. A month is a useful place to begin if your expense and customer records are available by month. Use the same dates for the spending and the new-customer count. If you include a bill from a different period, write that down so someone reading the result understands the mismatch.

Next, decide which costs belong in this version of CAC. A business-wide view may include ad spend and other sales or marketing expenses used to win new customers. You might also create a narrower paid-channel view that includes only selected campaign costs. Both can be useful, but they should not share a label that makes them look identical.

Finally, count customers, not orders. A person who buys twice is still one customer for a new-customer count. A purchase from someone who bought before may be useful revenue, but it is not a newly acquired customer for this calculation. Your records need a practical way to distinguish those groups. If they do not, mark the count as incomplete rather than making up a split.

The Business Development Bank of Canada defines CAC using the cost of gaining new customers and the number gained. The formula is a start. Use the costs and customer records from your own store.

Calculate the figure with a simple example

Imagine a small store reviews one month. It spent $3,000 on ads and another $1,500 on sales and marketing work that the team includes in this business-wide calculation. During the same month, it identified 100 new customers. The numbers are hypothetical and do not describe a KOLSprite customer or a typical store.

Add the included costs: $3,000 plus $1,500 is $4,500. Divide $4,500 by 100 new customers. The resulting CAC is $45 per new customer for that period and cost definition. It is an average, not the actual cost of finding each individual buyer.

Source: BDC, customer acquisition cost, accessed November 3, 2026. Dollar amounts and counts are made up to show the arithmetic.

The $45 figure changes if you change the inputs. If an expense covers work for both new and existing customers, decide whether and how to allocate it. Keep the basis of that allocation with the result. Someone else should be able to see what went into $4,500, what was left out, and how the 100 people were identified.

Why the ad dashboard can show a different cost

Now suppose the store's ad platform reports $3,000 in spend and 150 purchases for the same dates. The platform's cost per reported purchase is $20: $3,000 divided by 150. That is a valid calculation for those two values. It is not automatically the same as $45 CAC, because the platform's purchase count may include repeat customers and its spend figure does not include the other $1,500 in this example.

The counts may differ because a platform and your store use different rules. Each may count a sale over a different time span or assign it to a different source. Check the dates and event each report uses. Check how your store identifies a buyer, too. If two platforms report the same sale, adding their counts would not give you two distinct new customers.

This does not mean you must discard the $20 figure. It may help you watch a campaign under a consistent platform definition. The $45 business-wide figure may help you consider the total cost of gaining new customers over the period. The mistake is using one as the other without noticing that the numerator and denominator have changed.

If you sell on Amazon as well as your own store, review what off-site attribution reports before you compare the figures. That report has its own scope. It does not show the full cost of gaining every customer across your business.

Make the calculation useful for a decision

Write a short note next to the result. Include the dates, costs, new-customer count, and how you identified those buyers. If a cost covers both new and existing buyers, explain how you split it. Say which data is missing. The note keeps the number tied to the choices behind it.

Now decide what you need to know. Can the business afford to gain customers at this cost? Compare it with what a new buyer's order earns after the relevant costs. Add later value only when your records support it; do not assume a buyer will return to make the first order look profitable. For a choice between ads, use the same campaign measure for each and remember that it is not company-wide CAC.

A lower CAC is not a complete business goal on its own. A store can lower spending and also reach fewer new customers. It can attract more buyers while spending more per buyer. Compare the cost with the number of new customers and the business outcome you care about. A single average does not tell you whether every channel, product, or customer group behaved the same way.

For a fuller unit-cost view, check the costs attached to an order as well. Order margin and CAC answer different questions. Keep them separate before considering them together; otherwise you can count the same cost twice or leave an expense out of both views.

When your records cannot yet support CAC

Suppose you know ad spend but cannot tell new buyers from returning buyers. Call the result cost per purchase, not CAC. Your next task may be to improve the buyer count, while following your data and privacy rules. Until then, use the partial measure with its limits made clear. It is better than a precise CAC you cannot support.

Decide how cancelled and refunded first orders affect your count. Your records may treat these cases differently from an ad platform. Record the rule you choose, apply it consistently, and note whether a person who later places another order is counted once or more than once.

Keep the count of new buyers next to the CAC figure each month. A change from 100 to 50 new buyers matters even if the average cost stays at $45.

The same issue can arise in reverse: you know the buyer count but lack some costs. List what you included and what is missing. A paid-media measure can still guide an ad review if the team knows its limits. Recheck the figure when the missing costs are ready, using the same dates.

Before you compare months, check whether you counted the same costs and buyers. A new agency fee can move the figure. So can a new way of assigning sales or defining a first-time buyer. Note those changes beside the result; do not assume one campaign caused the difference.

For the next creative decision, define what a paid video test should prove.