Customer Acquisition Cost for Ecommerce: Blended vs Paid

Your best acquisition number might be a compliment.

The customer acquisition cost ecommerce founders quote is usually the blended one: all marketing spend divided by all new customers. That includes the buyer whose sister told her about you. You paid nothing for her, and she drags your average down.

Paid acquisition cost counts only the new customers your ads brought in. That's what the ads cost.

So use blended to judge the business. Use paid to set the ad budget. Mix them up and you can scale straight into a loss while your dashboard stays green.

Blended CAC is total marketing spend divided by all new customers, paid and organic. Paid CAC is paid spend divided by only the new customers those ads brought in. A store spending $52,000 for 1,400 new customers has a $37.14 blended CAC, even if its ads pay $80 per customer.

I've sat in that meeting, proud of a number that was mostly word of mouth. It feels great right up until you double the budget.

Why does every report show a different CAC?

Because each report is answering a different question, and nobody labels which one.

Run the math on a store like this, a hypothetical. A Shopify brand selling $148 enameled Dutch ovens, one month. Total marketing spend is $52,000: $48,000 on paid (media, agency, creative) and $4,000 on email and content. Shopify shows 1,400 first-time buyers. Meta and Google together claim 870 of them. The store's own order data traces 600 to a paid click.

The number	The math	What it answers
Blended CAC	$52,000 ÷ 1,400 = $37.14	Is the whole business efficient?
Platform CAC	$48,000 ÷ 870 = $55.17	What do the ad platforms say they did?
Paid CAC	$48,000 ÷ 600 = $80.00	What do ads cost per new customer?
Incremental CAC	$48,000 ÷ 420 = $114.29	What did the ads cause? (assumes a test shows 420)
Next-customer CAC	$12,000 ÷ 90 = $133.33	What happens if I spend more?

Blended $37.14

Platform reported $55.17

Paid $80

Incremental $114.29

Next 90 customers $133.33

Same Dutch oven store, same month. The first order leaves $72.64 of contribution to pay for acquisition.

The platform number is low for a boring reason. One buyer can get claimed twice. One analysis of platform numbers against Shopify puts it plainly: "Shopify only records each order once." That's why the bottom of your fraction should come from your store.

How do you calculate CAC, blended and paid?

Two formulas. Same dates on both sides.

Blended CAC = all marketing spend ÷ all new customers

Paid CAC = paid acquisition spend ÷ new customers from paid channels

Here's what goes where. Eightx's blended vs paid breakdown counts media, agency, creative, affiliate and paid influencer costs in the paid number, and I agree with it. If you only count media, you've built a third number and given it a flattering name.

Cost	Blended	Paid
Meta, Google, TikTok media	Yes	Yes
Agency or media buyer fees	Yes	Yes
Ad creative and paid influencers	Yes	Yes
Affiliate commissions	Yes	Yes
Email platform and content	Yes	No
Buyers from organic, referral, email	Counted	Not counted

The one rule founders break most: first-time buyers only. Shopify's New vs returning customers report gives you that split without a spreadsheet. Count repeat orders as customers and your CAC falls overnight, and you didn't do anything.

Why does blended CAC flatter you?

Here's the math on the Dutch oven. A $148 order, minus $46 landed cost, $19 shipping, $4.44 in payment fees (3%) and $5.92 in returns (4%). That leaves $72.64 of contribution on the first order. If you've never split that number out, the contribution margin guide walks each line.

Now put $72.64 next to the two numbers.

Blended says you pay $37.14 and keep $35.50. Beautiful.

Paid says you pay $80 and lose $7.36. On 600 customers, that's $43,584 of contribution against $48,000 of spend, so the ads lost $4,416 on the first order.

And the store still made money. The 800 organic buyers produced $58,112 on $4,000 of spend, which leaves $54,112. Add the two and you get $49,696.

Your word of mouth is paying your ad bill, and the blended number is what hides the transfer.

Now watch what happens when the founder sees that $37.14 and says, "Let's spend more." An extra $12,000 brings 90 more customers. Blended goes from $37.14 to $42.95 ($64,000 ÷ 1,490). That still looks wonderful. But each of those 90 cost $133.33, and they brought in $6,538 of contribution for $12,000. The new money lost $5,462 and the headline number barely blinked.

This is the same trap as a healthy marketing efficiency ratio that the bank account disagrees with. Averages move slowly. Bleeding starts at the edge.

How big is the gap for most stores? Eightx built an indicative table by vertical, with blended sitting about 10% below paid in electronics and about 35% below in food and beverage. They say up front that no public vendor reports this ratio cleanly, so it's a synthesis. Our Dutch oven store runs wider, at 54%, because I gave it a loud word-of-mouth base. The bigger your organic base, the louder the compliment.

Can you trust paid CAC either?

Not fully. And I'd rather you hear that from me than from your first holdout test.

Paid CAC can miss the buyer who saw an ad and later typed your name into Google. It can also credit ads for buyers who were coming anyway. Haus shows how much that matters in its Meta testing guide: in its example, Meta reports 100 conversions and 60 are incremental.

Apply that kind of haircut and the hypothetical store's 600 might be 420, which is where the $114.29 came from. That's an assumption until you test it. Meta's Conversion Lift has a seven-day minimum and needs at least 10% of your audience in each cell, and it needs enough conversions to read. Google also publishes Meridian GeoX, an open-source tool for geo experiments, though it's built around notebooks and an API, so it suits a store with an analyst.

And here's where automation changes your math. Meta's newer Advantage+ campaigns no longer accept the old setting that capped spend on existing customers. Its docs now say to build two ad sets, one for existing customers and one that excludes them. If you haven't split them, check who your budget is reaching, because a repeat buyer in the platform's report never shows up as a new customer in yours.

What is a good customer acquisition cost for ecommerce?

One your margin can pay for. I know that sounds like a dodge. Here's the number.

Polar publishes directional 2026 ranges from the brands it works with: $40 to $120 for paid social, $50 to $150 for non-brand search, $5 to $30 for branded search. Useful for spotting something broken. Useless for setting your budget, because $80 kills the Dutch oven on the first order and might be cheap for a $400 mattress.

So start with first-order contribution. Below it, you're profitable the day they buy. Above it, you're lending money to the customer, and repeat orders have to pay you back. That can be a great loan, and our LTV to CAC guide shows how to measure the payback, but you should know you're making it. A subscription is the same loan on a longer schedule, and the subscription vs one-time purchase math shows when it pays.

And cheap isn't the goal. Look at an animal repellent product we moved from Amazon to Shopify. Advertising per sale went from $2.69 to $26.36, nearly 10 times more. Average order value went from $17.95 to $101.38, and net profit per sale, after paying for ads and inventory, went from $3.83 to $48.27. See the full case study numbers. Real client numbers, not typical results, and not a promise of what your store will do.

A $2.69 ad cost that leaves you $3.83 is a worse business than a $26.36 ad cost that leaves you $48.27.

How do you lower paid CAC without touching the ads?

Paid CAC is two things. Cost per visitor divided by the share of visitors who become new customers.

The Dutch oven's $48,000 bought 40,000 paid sessions, so $1.20 each. Conversion rate to a first order was 1.5%. $1.20 ÷ 1.5% = $80. Everybody works on the $1.20. Your media buyer sweats it for pennies. And fixing creative fatigue is worth doing when it's the real cause.

Now hold the $1.20 still and fix the page. Say the conversion rate goes to 2.1% and the order grows to $171 because the page finally sells the lid and trivet set. Conversion rate 1.5%, average order value $148: revenue per visitor $2.22, or $22,200 on 10,000 visitors. Conversion rate 2.1%, average order value $171: revenue per visitor $3.59, or $35,910 on 10,000 visitors.

Paid CAC drops from $80 to $57.14 ($48,000 ÷ 840). Contribution per order rises to $84.03, with $9 more product cost, $1 more shipping, and fees and returns on the bigger order. The ads go from losing $4,416 to making $22,585. Same spend. Same clicks.

This is the part AI has changed for a small team. Pulling the inputs got easier: Shopify's Sidekick can build a report from a plain question, and Shopify says its channel marketing reports cover customer acquisition cost. Read how any report defines it before you trust it. And the page got faster: AI can now draft the headline, the objection copy and the bundle around the question a buyer asks before paying $148 for a pot. Will this thing outlive the cheap one I already own? Answer that and you've cut paid CAC without bidding on anything.

Once paid CAC sits below contribution with room to spare, that's when to scale Facebook ads. And track it where you track everything else, in a weekly profit view.

What to do next

Take last month. Divide your paid spend, fees included, by first-time buyers from paid channels. Write that number next to your first-order contribution. If paid CAC is higher, freeze the budget increase and look at the page those ads land on first.

---

Book Your Profit Audit

If your blended number looks great and your bank account doesn't agree, the gap usually lives between the ad and the checkout. On a free profit audit we find your paid acquisition cost, show you what each visitor is worth, and show you how to rebuild a high-converting product sales page in less than 15 minutes.

Book Your Profit Audit →

Or go here to check it out → revenueflows.ai

P.S. Blended CAC is the compliment. Paid CAC is the invoice. Pay attention to the one that comes with a due date.