Common CAC Calculation Mistakes (and How to Fix Them in Your CAC Formula)
by Trivas.ai
|
6 min read
Oct 14, 2026
CAC looks like the easiest metric in ecommerce. Spend divided by new customers. Two numbers, one division. And yet it's one of the most mangled numbers in any DTC dashboard, because the two inputs are rarely as clean as they look.
Most teams aren't lying about CAC. They're just quietly inconsistent: pulling spend from one source, customers from another, and never checking whether the two actually line up. The common CAC calculation mistakes in this post show up at brands of every size, from scrappy seven-figure Shopify stores to Amazon sellers running six-figure monthly ad budgets.
Quick note before we get into it: we're folding this page into a broader CAC guide soon, since "CAC calculation mistakes" on its own doesn't pull enough search volume to justify a standalone page forever. The content isn't going anywhere, it's just getting a better home inside our guides and reports library.
Here's why it matters regardless of where the page lives: a wrong CAC number doesn't just look bad in a slide deck. It drives real decisions, like which channel gets next month's budget and which one gets cut. Get the formula wrong and you'll defund something that was actually working, or keep funding something that wasn't.
Mistake 1: Mixing Blended and Paid CAC
Blended CAC is total marketing spend divided by total new customers, full stop. Paid CAC is ad spend only, divided by customers who came through paid channels. They're both valid metrics. The mistake is using them interchangeably, or worse, reporting one while making decisions based on the assumption it's the other.
Say you spent $50,000 total last month, $30,000 of which went to paid ads, and brought in 500 new customers. Blended CAC comes out to $100. Paid CAC comes out to $60. If your performance marketers are judging a specific campaign against the $100 blended figure, every paid channel looks worse than it actually is, and you'll end up cutting budget from something that's clearing your profitability bar just fine.
The fix isn't picking one metric and ditching the other. It's labeling them clearly every single time, and never letting blended CAC sneak into a conversation about paid-channel performance.
Mistake 2: Ignoring Attribution Window Mismatches
This one's sneakier because it doesn't show up as an obvious error, just a number that feels slightly off.
Meta Ads Manager defaults to a 7-day click attribution window. GA4 and Shopify often run on 30-day windows by default. If you pull spend from Meta's dashboard and new-customer counts from Shopify, you're comparing two different definitions of "who converted" without realizing it. Depending on the direction of the mismatch, CAC gets inflated or deflated, and nobody notices because the number still looks plausible.
The fix is boring but it works: pick one attribution window, apply it everywhere, and stop letting each platform report on its own terms. It doesn't have to be the "right" window, there isn't really such a thing. It just has to be the same window across every channel you're comparing.
A lot of CAC formulas quietly only count spend tied to "new customer" campaigns, and leave out retargeting and retention spend entirely. That's a problem the second your remarketing budget gets substantial, which for most mature DTC brands, it does.
If 20% of your monthly ad spend goes to retargeting past visitors and email subscribers, and none of that spend is in your CAC numerator, your "new customer CAC" is going to look better than reality. You're acquiring customers with help from that retargeting spend, you're just not counting it.
The cleaner approach is to stop trying to cram everything into one blended number. Track new-customer CAC and total marketing cost per order (or per revenue dollar) as two separate metrics. They answer different questions, and forcing them into one figure just hides the tradeoff.
Ad spend is the easy part to track because it's sitting right there in the platform. The costs that get left out are the ones spread across invoices nobody's tying back to acquisition:
Agency retainers for media buying
Creative production (photo, video, UGC fees)
Software and tool spend used specifically for acquisition (landing page builders, creative tools, tracking software)
Affiliate and influencer commissions
Leave these out and CAC looks artificially healthy. That's not a rounding error, either. A brand spending $8,000/month on an agency retainer plus $3,000 on creative production, on top of $40,000 in media spend, is understating CAC by more than 25% if it only counts the media line.
The rule of thumb: if the dollar was spent to acquire the customer, it goes in the numerator. Doesn't matter if it was paid to Meta or to a freelancer on Upwork.
Mistake 5: Comparing CAC Across Channels Without Normalizing Timeframes
Comparing this month's Meta CAC to last quarter's Google CAC tells you almost nothing, but it's one of the common CAC calculation mistakes that happens constantly because the numbers are just sitting there in two different dashboards, begging to be compared.
Seasonality moves CAC on its own. So does a platform algorithm update, an iOS tracking change, or a competitor's price drop. Comparing across mismatched timeframes bakes all of that noise into a conclusion that looks like it's about channel performance, when it's really about timing.
Compare CAC on matching windows instead, same week against same week, same month against same month. And treat any single-point CAC reading with suspicion. A trend line across 8 to 12 weeks tells you far more than one snapshot ever will, because it shows you whether CAC is actually moving in a direction or just bouncing around normal week-to-week noise.
Getting CAC Right Without the Manual Spreadsheet Work
Every mistake above comes down to the same root cause: pulling numbers from different systems that don't agree on definitions, then stitching them together by hand and hoping nothing slipped through.
That's genuinely hard to avoid in a spreadsheet workflow. It's a lot easier to avoid when spend and order data from Amazon, Shopify, Meta, and Google all land in one Redshift-backed warehouse with consistent attribution logic applied across the board, which is what BI reporting is built to do. Window mismatches and missing cost lines stop being a weekly hunt because the data's reconciled before it ever hits your dashboard.
The AI Wingman layer on top of that goes a step further, flagging CAC anomalies as they happen instead of leaving someone to notice three weeks later that paid CAC quietly jumped 40%.
If you want to see what consistent, cross-channel CAC actually looks like instead of a patchwork of spreadsheet formulas, take it for a spin with a free trial. Worth seeing the number you can actually trust for once.
Content author and contributor at Trivas.ai, sharing insights on e-commerce analytics, business intelligence, and data-driven strategies to help businesses grow.
Continue Reading
explore more insights
Machine Learning in E-Commerce: Transforming Retail with Intelligent Insights
3 min read
What CMOs Should Know About Ecommerce Analytics (Before the Next Board Meeting)
.d53b12e5.png&w=3840&q=75 "Trivas.ai")
