Guide · Ecommerce economics
What customer acquisitionactually costs by category
Published CAC figures for the same category in the same year differ by three times. That is not sloppy research, it is three different denominators being given the same name. Here is what each one measures, what the ranges actually are, and how to work out the number your own margin can carry.
01 · Why the numbers disagree
Three different questions,one word
Search for beauty CAC and you will find $42 and $110 published in the same year, both sourced, both defensible. Fashion comes back at $37 and at $90. The published spread inside a single category is wider than the spread between categories, which makes an unqualified benchmark almost useless as a diagnostic.
The reason is the denominator. Three things get called customer acquisition cost.
All three are correct answers to different questions, and a benchmark article that does not say which one it used is not comparable to your account. Shopify's own 2026 commerce reporting puts a merchant-wide average near $318 once every acquisition cost is counted, which is a fourth denominator again.
Before comparing yourself to anything, write down what is in your numerator and what is in your denominator.
Our own rule is simple. Paid spend plus everything spent to make the ads work, divided by first-time buyers only, measured monthly. It produces an unflattering number, which is the point. If a channel works at that number it works, and the platform's version will always look better.
02 · The ranges
CAC and order value,by category
Published 2026 figures. Where two credible sources disagree, both ends are shown rather than averaged, because the disagreement is the useful information.
| Category | Published CAC | Order value | Gross margin | Returns | Repeat, 12mo |
|---|---|---|---|---|---|
| Beauty & personal care | $42–110 | $55–137 | 60–70% | 4–12% | 25–40% |
| Supplements & health | $71–89 | $45–120 | 65–78% | — | 30–45% |
| Fashion & apparel | $37–90 | $80–200 | 50–60% | 20–40% | 20–32% |
| Food & beverage | $51–75 | $45–147 | 40–55% | — | 30–45% |
| Pet | ~$23 | $55–110 | — | — | 30–35% |
| Home & furniture | ~$45 | $95–295 | — | under 18% | under 18% |
| Electronics | Above average | $120–348 | Thin | 8–15% | Low |
| Jewelry & luxury | $91–400 | $180–436 | — | ~8% | ~10% |
A dash means we could not find that figure published consistently enough to print. That is a real answer rather than an omission, and it is exactly why the tool below asks for your own margin instead of assuming one. For context, the median DTC gross margin across eleven public filings is 57 percent, with the middle half between 46 and 64 percent. The blended ecommerce averages of $68 to $84 sit inside this table and describe almost nobody in it.
03 · Free, no email required
The only benchmarkthat matters is yours
Pick a category to load its published bands, then overwrite anything you know. This works out what one customer can cost you before the trade stops making sense, and puts that ceiling next to what the category is actually paying.
Your economics
Loaded: beauty & personal care, published midpoints.
Repeat rate is doing more work here than anything else. It is the difference between a ceiling set by one order and a ceiling set by a customer. Where a category cell was blank in the table above, this loads the 57% median DTC gross margin and the 19% overall return rate instead, so overwrite those first.
What you can afford
Updates as you type. Nothing is sent anywhere.
Then go and set the target
This page gives you a ceiling. The max CPL calculator turns it into the bid target, cost per lead and blended figures you actually work to, with an ecommerce mode built for exactly this.
04 · The other axis
The same customer,priced by channel
Category sets what a customer is worth. Channel sets what they cost. Published 2026 channel-level acquisition costs, which are not directly comparable to each other and should not be added up.
| Channel | Published CAC | What to know about the figure |
|---|---|---|
| Meta | $212–230 | Channel-level estimates with repeat buyers stripped out. The same account reports a median cost per purchase near $39 in its own interface, and both numbers describe the same spend. |
| Google Ads | $50–130 | Wide because it blends Shopping, brand search and non-brand search. Strip your own brand terms out before you believe the low end. |
| TikTok | $90–129 | Cheaper impressions than Meta, lower purchase intent. The creative does not transfer, so the real cost includes a second production pipeline. |
| Influencer | $40–300+ | The widest band on the list, driven almost entirely by creator tier and whether the content is licensed for paid amplification afterwards. |
| Referral | $40–65 | Cheap and capacity-limited. It scales with your customer count rather than your budget, which makes it a consequence of retention rather than a channel. |
| Email and SMS | Near zero | Marginal cost only. The acquisition already happened somewhere else, which is why retention lowers blended CAC without touching a single ad account. |
| Organic search | $200–300 | Fully loaded and amortised, which is why the figure looks alarming next to a free channel. It falls over time while paid channels rise. |
Two structural pressures sit under all of these. Google search click costs rose roughly 13 percent year on year and Meta CPMs around 20 percent, and the signal loss that began with Apple's tracking prompt never fully reversed. Nobody is winning this auction by bidding better. They are winning it by being able to pay more, which is what the rest of this page is about.
05 · The reframe
Your category does notset your CAC ceiling
Two supplement brands with identical $89 acquisition costs can be in completely different businesses. One keeps 70 percent after cost of goods and gets a second order from four in ten customers. The other keeps 45 percent, sells a one-time product, and is quietly insolvent at the same number.
What actually sets the ceiling is contribution margin multiplied by expected orders. Everything else is commentary.
Contribution margin is what survives after cost of goods, shipping, payment and platform fees, and the orders that come back. Across DTC it has fallen from roughly 35 percent in 2021 to about 22 percent in 2025, mostly because acquisition got more expensive. A workable, scalable contribution margin in DTC is around 20 percent, and gross margin above 60 percent is what strong brands target, with anything under 45 percent a warning sign.
Every point of contribution margin raises what you are allowed to pay for a customer. So does every point of repeat rate.
That is the whole strategic argument for spending on conversion rate and retention before spending more on ads. A brand that lifts its store conversion rate from 1.5 to 2 percent has not saved money, it has raised its bid ceiling by a third. A brand that lifts repeat rate from 20 to 30 percent has done the same thing again, and neither move required outbidding anyone.
It also explains the pattern in the category table that looks strange at first glance. Pet products carry the lowest published CAC at around $23 and jewelry the highest at up to $400, and both can be good businesses. Pet wins on frequency, luxury wins on margin, and the categories that struggle are the ones in the middle with thin margins and no reason to reorder.
06 · The rule everyone quotes
3:1 came from software,not from retail
The 3:1 lifetime value to acquisition cost ratio is the most repeated number in ecommerce and the least examined. It came out of subscription software, where revenue arrives monthly, churn is measurable and the lifetime is genuinely predictable. None of those things is reliably true of a store selling physical goods.
Published guidance still puts 3:1 as a minimum, with most DTC brands targeting 3:1 to 4:1 and anything above 5:1 read as underspending on growth. Those are reasonable rules of thumb. The problem is that the same ratio can describe two completely different cash positions.
A brand at 3:1 where the payback lands on the first order is printing money. A brand at 3:1 where the payback lands eighteen months out is financing its own growth with working capital it may not have. The ratio is identical. The businesses are not.
So the number worth managing is payback period, not the ratio. Under six months is the commonly cited target, and under 90 days is where a brand can genuinely fund next month's acquisition out of last month's customers rather than out of the balance sheet.
That is also the honest reason the ceiling in the tool above is shown twice. The first-order figure is what you can pay and stay liquid. The twelve-month figure is what you can pay if you are willing to fund the gap, and whether you should depends on your cash position rather than your marketing plan.
One more sanity check that costs nothing: median DTC net margin runs 3 to 10 percent, and median EBITDA has compressed to roughly 7 to 8 percent. If your acquisition maths only works on an optimistic lifetime assumption, the category is telling you it usually does not.
Common questions
Straight answers
What is a good CAC for ecommerce?
There is no answer to that question without your margin and your repeat rate. Published blended averages sit at $68 to $84, and category figures run from about $23 in pet products to $400 in luxury, but a good CAC is defined by what a customer contributes rather than by what other people are paying.
The workable version: your acquisition cost should sit below your contribution over the payback window you can afford to fund. The tool on this page works that out, and the max CPL calculator turns it into a target.
Why does our CAC look so different from the benchmarks?
Almost always the denominator. Platform cost per purchase, blended CAC and paid new-customer CAC are three different calculations that all get called CAC, and they can differ by a factor of six on the same account in the same month.
Check whether the benchmark counted repeat buyers, whether it included agency and creative costs, and whether it used platform-attributed conversions or actual new customers. If the source does not say, it is not comparable.
Should we use blended CAC or paid CAC?
Both, for different decisions. Blended CAC tells you whether the business is working, since it includes the organic and referral customers your paid spend partly creates. Paid CAC tells you whether the next dollar of ad spend is worth spending.
Managing only to blended hides a failing ad account behind a healthy retention programme. Managing only to paid ignores the halo the ads produce elsewhere. Report both and never mix them in one sentence.
Our CAC keeps rising. Is that us or the market?
Partly the market. Meta CPMs rose roughly 20 percent year on year and Google search click costs around 13 percent, and the targeting signal lost after Apple's tracking prompt never fully returned. Everyone in your auction is paying more.
What is inside your control is the ceiling rather than the cost. Conversion rate, contribution margin and repeat rate all raise what you can afford to pay, and they are the only durable way to win an auction that gets more expensive every year.
How do we lower CAC without cutting spend?
Mostly you do not lower it, you raise what you can pay. But two things genuinely reduce blended CAC without touching the ad account.
The first is retention, because email and SMS revenue arrives at near-zero marginal cost and dilutes the blended figure. The second is stripping your own brand terms out of prospecting campaigns, since paying to acquire customers who already searched for you inflates both the spend and the apparent result.
Where did these figures come from?
Published 2026 benchmark studies and public filings, listed with their scope at the bottom of this page. Where sources disagreed we printed both ends of the range rather than averaging them, and where no figure was published consistently we left the cell blank instead of estimating.
None of it is client data. We do not publish client numbers as benchmarks, because a handful of accounts is an anecdote rather than a dataset.
Methodology
Sources
Every figure on this page traces to one of these. Scope is stated because scope is what makes a benchmark usable.
- Category CAC ranges. 2026 DTC vertical benchmark compilations drawing on First Page Sage client data and aggregated agency datasets covering brands in the $5M to $75M range, alongside separate 2026 CAC statistics roundups. These are the sources that disagree with each other, which is discussed above rather than smoothed over.
- Average order value by vertical. 2026 AOV benchmark set across ten ecommerce verticals, including the DTC Shopify average of $85 to $95 and the Amazon marketplace average of $52.
- Gross margin. Median DTC gross margin of 57 percent derived from eleven public company filings on SEC EDGAR, with the middle half between 46 and 64 percent, plus category-level ranges from 2026 profitability benchmark work.
- Contribution margin. Median DTC contribution margin falling from roughly 35 percent in 2021 to about 22 percent in 2025, per Fairview data cited in 2026 D2C benchmark reporting.
- Return rates. National Retail Federation 2025 Retail Returns Landscape for the overall ecommerce figure, with category ranges from 2026 returns-industry benchmark datasets.
- Repeat purchase rates. Twelve-month DTC benchmarks of 25 to 30 percent from industry aggregators, alongside a February 2026 analysis of 156,110 DTC customers that put the aggregate at 18.8 percent. Both are quoted because the measurement window explains the gap.
- Channel-level acquisition costs. 2026 channel CAC benchmark work covering Meta, Google, TikTok, influencer, referral and organic, with the fully loaded organic figure noted as amortised.
- Auction inflation. Published 2026 platform benchmark reporting on Meta CPM and Google cost-per-click year-on-year movement, and Triple Whale's ecommerce Meta benchmark covering August 2025 to July 2026.
Benchmarks age. This page is dated at the top and we refresh it when the underlying studies update. If you find a figure here that a newer study contradicts, we would rather hear about it than not.
What to do with this
Set the ceiling,
then go shopping.
The number that matters is the one your own margin produces. Work it out, turn it into a bid target, and only then argue about which channel to spend it in.