eCommerce
Amazon vs DTC margin benchmarks: where your SKU nets more
On the same product, DTC typically nets about 7 points more contribution margin than Amazon FBA, roughly 36% versus 29% on a $50 order. But below about $45-50 average order value, Amazon wins, because flat DTC shipping and customer acquisition cost eat too much of a small order.
Key Takeaways
- On a $50 order, DTC keeps about 36% contribution margin versus 29% on Amazon FBA, a 7-point gap driven mostly by Amazon's 15% referral fee against a 3% DTC payment fee. The gross margin is identical; the channel cost is not.
- Amazon's effective take rate now exceeds 50% of seller revenue for most categories once referral, FBA, storage, and advertising stack up. 49% of active sellers cite fees as their top margin concern and 47% saw margins decline year over year.
- The crossover sits near $45-50 average order value, but it moves with your CAC. An organic-heavy brand can push it down to $25-30. A CAC-heavy brand may not cross over until $70 or more. This is the single most volatile input in the model.
- Amazon ads cost $12-18 per conversion against $38-58 DTC paid CAC. That 2-3x gap is the real reason brands tolerate Amazon's lower gross margin: the marketplace substitutes cheaper traffic for the referral premium.
- Category beats channel. Beauty wins on both sides (51.9% Amazon gross margin, mid-to-high-teens DTC CM3). Home goods collapse to low single-digit DTC margin because bulky shipping eats 15-25% of the product price, which counterintuitively makes Amazon the better home for some heavy SKUs.
If you run a brand on both Amazon and your own Shopify store, you have probably stared at two revenue numbers and assumed the bigger one is the better channel. That is the wrong question. The question that matters in 2026 is which channel nets more margin per dollar of inventory you deploy, because that is the number that compounds. The short version: on the same product, DTC usually wins per order, but not always, and the place it flips is lower than most operators expect. Here is the SKU-level math, what to watch as Amazon fees keep climbing, and how to split your inventory so each unit lands in its most profitable channel.
The framing trap is treating Amazon's cut as a cost of goods. When I talk to founders running a brand this size, the first thing I tell them is to stop burying Amazon fees inside COGS. The referral fee, FBA warehousing, storage, all of it: those are channel costs, not product costs. Your gross margin on a given SKU is identical whether it sells on Amazon or your site. What differs is contribution margin, after each channel takes its cut. Get that separation right and the comparison gets honest fast.
The real Amazon fee stack: what 50% of revenue pays for
Amazon's headline referral fee is the part everyone knows, but it is the smallest layer of the stack for most brands. The referral fee is a percentage of selling price that varies by category, and for apparel it has a step structure that quietly punishes higher prices: 5% on items under $15, 10% from $15 to $20, then 17% above $20. Home and kitchen is a flat 15%. Beauty is 8% under $10 and 15% above. On top of that sits FBA fulfillment (a per-unit fee tied to size and weight), monthly storage, inbound placement, returns processing, and advertising.
| Category | Referral structure | FBA per unit (est.) | Storage / inbound | Total non-ad fees |
|---|---|---|---|---|
| Apparel (under $15) | 5% under $15 | $3-5 | ~2% | ~25-30% of price |
| Apparel ($15-20) | 10% from $15-20 | $4-6 | ~2% | ~35-40% |
| Apparel (over $20) | 17% above $20 | $5-8 | ~2% | ~40-50% |
| Beauty (under $10) | 8% under $10 | $3-4 | ~2% | ~25-30% |
| Beauty (over $10) | 15% above $10 | $4-6 | ~2% | ~35-40% |
| Home & kitchen | Flat 15% | $4-8 | ~2% | ~35-45% |
Add advertising and the picture gets worse. Marketplace Pulse called it back in 2023: Amazon's effective take across fees and ads now exceeds 50% of seller revenue for most categories, up from roughly 40% five years earlier, and the trend has held through 2026. In Jungle Scout's seller survey, 49% of active sellers name fees as their primary margin concern, 46% name advertising, and 47% saw margins decline year over year. Amazon froze referral rates in 2025 and 2026 but added a 3.5% fuel and logistics surcharge on FBA fulfillment fees from April 2026, so the standard-size fee load is still drifting up. When you model an Amazon SKU, "gross revenue" is a vanity number. Model the take rate.
The DTC cost stack: why your per-order margin is higher on Shopify
Your own store has a thinner, but not free, variable stack. Payment processing runs 2-4%. Shipping and fulfillment land 8-12% of revenue for most brands, more if your product is heavy. Platform and app fees take 1%. Returns provision sits around 2% for most categories. And then there is the one input that swings the whole comparison: customer acquisition cost.
This is where the CM1-versus-CM3 framework earns its keep. CM1 is contribution margin before marketing (revenue minus COGS, fulfillment, shipping, payments, returns). CM3 is after you load in blended ad spend. The gap between them is your CAC efficiency, and it is enormous. A brand with strong email, SMS, and organic might run blended CAC at 10-12% of revenue. A brand buying most of its growth on Meta might run 25-30%. Same product, same COGS, wildly different CM3, and that spread is exactly what separates Amazon and Shopify contribution margins across verticals.
Operators feel this asymmetry most on the data side. The thing founders keep telling us they hate about Amazon is the lack of visibility. You cannot see lifetime value, because Amazon keeps the customer. On Shopify you can, in theory, follow a buyer across their whole relationship with you. The catch: the moment that same buyer reorders on Amazon for the next-day convenience, you lose the thread. If you could perfectly track LTV across both channels, the "which is more profitable" question would look very different per customer than it does per order.
Returns are quietly eating your margin. See by how much.
Get our Real Cost of Returns calculator: plug in your numbers, see the true hit per return.
Check your inbox. We'll send the Real Cost of Returns calculator shortly.
Same SKU, two channels: the side-by-side
Here is the comparison that matters, on a $50 order with 30% COGS for a mature brand (not launch phase). Amazon carries a 15% referral, 10% FBA, 2% storage and inbound, 2% returns, and 12% TACoS. DTC carries 3% payments, 10% shipping and fulfillment, 1% platform, 2% returns, and 18% blended CAC.
The COGS layer is identical, as it should be. The divergence is almost entirely the referral fee against the payment fee: Amazon's 15% cut versus your 3%. DTC spends more on advertising in this scenario (18% versus 12%), which narrows the gap, but the owned channel still lands at roughly 36% contribution margin against Amazon's 29%. That 7-point spread is the structural per-order advantage of selling on your own store at this price point.
But the spread is not uniform across categories, because gross margin and return rates are not uniform. Beauty is the standout on both sides. Apparel and home goods get punished, for different reasons.
| Category | Amazon gross margin | Amazon net (SDE) | DTC CM3 after CAC | DTC return rate |
|---|---|---|---|---|
| Beauty & personal care | 51.9% | 41.3% | Mid-to-high teens % | 1-5% |
| Health & personal care | 49.1% | 39.8% | Low-to-mid teens % | Low |
| Home & garden | 42.9% | 33.9% | Low single digits % | 15-20% |
| Apparel | Not separately published | n/a | Low-to-mid teens % | 25% (range 20-40%) |
| All FBA average | 45.6% | 36.9% | n/a | 14% overall |
The home-goods line is the one that surprises people. DTC contribution margin after CAC collapses to low single digits, not the teens, because bulky-unit outbound shipping eats 15-25% of the product price before you have spent a dollar on ads. That is the counterintuitive result: for some heavy home SKUs, Amazon's referral fee is actually cheaper than what it costs you to ship the thing yourself. Category determines the winner more often than channel does.
The break-even AOV: when Amazon wins the math
The 7-point DTC advantage at $50 is not a constant. It is a snapshot on a curve. Amazon's FBA fee is largely fixed per unit, so on a cheap order it is a big share of revenue and on an expensive order it is a small one. DTC's flat shipping behaves the same way, but DTC also carries CAC as a percentage of order value, which compounds against small orders.
Scenario note: This curve uses flat-dollar fulfillment inputs ($4.50 FBA, $7 flat DTC shipping) and a 15% blended DTC CAC as a share of each order value. It is a lower-CAC model designed to isolate how margin moves with price across a wide AOV range. The $50 cost-stack chart above uses %-based fulfillment and an 18% CAC rate, which is why DTC shows 36% versus 29% at $50 there rather than the ~26% versus ~25% on this curve. The crossover shape and direction are the same in both models; the absolute margins differ because the scenarios differ.
Below roughly $45-50 order value, Amazon wins, because a $7 shipping label plus acquisition cost on a $25 order is brutal math. Above $45-50, DTC pulls ahead and keeps widening. The crossover is the decision point, and it is the most CAC-sensitive number in this entire post. Treat the $45-50 figure as illustrative, not gospel. A brand with cheap organic traffic shifts the crossover down to $25-30. A brand that buys most of its growth pushes it up past $70. Before you make any channel call, rebuild this curve with your own CAC and your own shipping cost.
The same $50 product nets about 36% on your store and 29% on Amazon. Flip the order value to $25 and Amazon wins. Flip your CAC from 12% to 28% and the crossover moves $40. There is no universal answer to "which channel is more profitable." There is only your SKU, your price, and your acquisition cost.
Why brands tolerate Amazon's lower margin
If DTC nets more per order at most price points, why does anyone keep selling on Amazon? Because the per-order margin is only half the equation. The other half is what it costs to get the order in the first place.
Amazon's per-conversion ad cost runs $12-18 across categories. DTC paid CAC on Meta and Google runs $38-58. Even after you blend in organic, DTC acquisition is materially more expensive than Amazon's Sponsored Ads. The marketplace is, in effect, selling you cheaper traffic in exchange for the referral fee. For a brand that has not yet built an owned audience, that trade can be worth it, because the alternative is paying triple the CPA to acquire a customer you might not retain.
This is also why consumables behave differently. When someone sells a highly consumable product, I usually tell them to just be on Amazon. When a buyer runs out of something, they do not want to wait five days. Amazon gets it there tomorrow, and you reach repeat buyers who would never have found your site. The convenience is the moat, and fighting it on a low-AOV refill product is a losing use of your DTC margin. For a category-level breakdown of how this plays out in practice, see our supplements Amazon vs DTC economics breakdown.
SKU-level allocation: the framework that actually works
The brands that win this do not pick a channel. They allocate by SKU. The working principle: Amazon is an acquisition surface, DTC is where you capture margin and retention. Three tiers fall out of that.
Tier A, DTC-first SKUs. Hero products with strong margin, higher AOV, and a real brand story. These belong on your site, where the 7-plus points of contribution margin compound and where you keep the customer relationship and the LTV data.
Tier B, Amazon-velocity SKUs. High-turnover commodity and consumable items with adequate margin. Let Amazon's cheaper traffic and next-day delivery do the work. Use these to acquire, then convert repeat buyers to your owned channel with inserts and QR codes that make reordering on your site easier than reordering on Amazon.
Tier C, pull candidates. Any SKU where the fee stack drives Amazon contribution near zero, usually heavy, low-ASP, or sub-$15 apparel items paying the full FBA load. Model these honestly and be willing to delist them from Amazon entirely.
We watched one brand around $25K a month deliberately walk its mix from 80% Amazon toward 60% over a year, not by cutting Amazon revenue but by building the DTC side underneath it. The Amazon fees were roughly 15% of price, Shopify shipping roughly 10-12%, so once we modeled contribution separately by channel, the case for moving hero SKUs to the owned store was obvious. One constraint to plan around: Amazon's pricing algorithm. You cannot sell cheaper on your own site than on Amazon without risking a flagged listing, so DTC-exclusive discounts and bundles have to be structured carefully to stay above your Amazon price.
One more pressure point for 2026. Amazon new seller launches hit a decade low in 2025, down 44% year over year, while fee layers (inbound placement, low-inventory surcharge, the new fuel surcharge, ad auction inflation) keep stacking. Established sellers who used to net 15-25% are increasingly netting 10-20%. The distribution below shows where sellers actually land.
| Net margin band | Share of sellers | Context |
|---|---|---|
| Negative (loss) | ~13-15% | New sellers, poor unit economics, heavy ad spend |
| 0-10% | ~20-25% | Break-even; at risk from any fee increase |
| 10-20% | ~40-50% | Modal band for established profitable FBA brands |
| 20-35% | ~20-25% | Strong niches: beauty, supplements, premium |
| 35%+ | ~5% | Exceptional economics, rare outside premium or niche |
The implication is straightforward. The math increasingly favors DTC for any brand with the ability to drive owned traffic, and increasingly favors Amazon only where convenience and cheap CPA genuinely win. The brands that run this at the SKU level, rather than by gut, are the ones building better blended portfolio margin. For a deeper build of your own channel model, our fractional CFO services page covers how we set these up.
Sources and methodology
Amazon fee-stack and take-rate data. Referral, FBA, and storage figures are compiled from Amazon Seller Central's published 2026 fee schedules as summarized in third-party fee analyses, including Feedvisor's referral-fee breakdown and dated 2025-2026 fee-update coverage. The "50%+ effective take rate" figure traces to Marketplace Pulse's analysis. Referral rates cited are US marketplace standard rates; FBA estimates assume standard-size items, and oversize or hazmat units show materially worse economics.
Amazon seller profitability and registration trends. Profitability bands, fee-concern shares, and the decade-low registration figure come from Jungle Scout's State of the Amazon Seller 2025 survey and Marketplace Pulse seller-economics reporting. Survey data is self-reported and subject to selection bias (profitable sellers respond more often), so treat the distribution as directional, not census-level.
Amazon gross-margin-by-category data. The 51.9% beauty and 45.6% all-category gross-margin figures are from FBA Guys' category analysis, based on single-category FBA sellers. These skew toward private-label sellers; brand-led multichannel sellers with pricing power may run modestly different economics.
Acquisition-cost benchmarks. Amazon CPA figures are from Triple Whale's Amazon Ads Benchmarks by Industry, based on Sponsored Ads conversion data across roughly 30,000 brands. DTC paid and blended CAC come from Polar Analytics' ecommerce benchmarks, drawn from a panel of several thousand Shopify brands using a layered CM1-CM3 model.
Worked-example assumptions. The $50 cost-stack and the AOV crossover curve are modeled scenarios using the COGS, fee, shipping, and CAC inputs stated in each section, synthesized from the fee analyses above and DTC contribution-margin benchmarks. They are illustrative of a mature, standard-size-product brand. The crossover point is highly sensitive to actual CAC and shipping cost and should be rebuilt on first-party numbers before any channel decision.
Limitations. Amazon does not publish per-seller or per-category margin data, and no public multichannel brand separately discloses Amazon-versus-DTC contribution margin in its filings (SEC EDGAR review confirmed no standalone Amazon P&L segment exists). All fee-stack figures are therefore synthesized from third-party analyses of public fee schedules. DTC CAC is the single most volatile input and will move the break-even AOV materially by brand.
Frequently asked questions
what is a typical contribution margin on amazon after all fees and advertising?
For an established FBA brand, plan on roughly 25-30% contribution margin on a mid-priced order after referral, FBA, storage, returns, and advertising. Net margin after fixed overhead usually lands 10-20%. Beauty and supplements run higher, apparel and bulky home goods run lower.
is selling on amazon more profitable than selling on my own website?
Per order, DTC usually nets more contribution margin on the same product, about 7 points more at a $50 order value, because you avoid Amazon's 15% referral fee. The exception is low order values and high DTC acquisition cost, where Amazon's cheaper traffic wins.
at what price point does dtc become more profitable than amazon per order?
The crossover sits near $45-50 average order value in a typical model, but it is highly sensitive to your CAC. An organic and email-heavy brand can cross over around $25-30. A paid-traffic-dependent brand may not cross until $70 or more. Run it on your own numbers, not the benchmark.
how much does amazon take out of every sale in total fees?
Once you stack referral (5-17%), FBA fulfillment, storage, returns, and advertising, Amazon's effective take exceeds 50% of revenue for most categories. That is why gross revenue on Amazon is a misleading number to plan against.
what is a good dtc contribution margin after cac in 2026?
Healthy DTC brands run 15-30% contribution margin after acquisition cost (CM3). Top-quartile brands hit 30-40%. Below 10% you are likely overspending on paid traffic or carrying shipping you have not engineered out, especially on heavy products.
why is my amazon margin lower than my dtc margin even though amazon has more volume?
Volume and margin are different problems. Amazon takes a referral cut on every unit that your own site does not, so each Amazon order nets less even when the channel ships more units. The fix is to treat Amazon as an acquisition surface and migrate repeat buyers to your owned channel where the per-order margin is higher.
how should i allocate inventory between amazon and my dtc website?
Run contribution margin per SKU per channel, not blended. Push hero and high-margin SKUs toward DTC, keep high-velocity commodity and consumable SKUs on Amazon, and pull any SKU off Amazon where the fee stack drives contribution near zero. It is a capital-allocation decision, not an either-or.
what percentage of amazon sellers are actually profitable?
Most are. Survey data shows 60%+ of sellers maintain margins above 10%, but only about one-third clear 20%, and the modal band is 10-20%. Margins have compressed since 2022-2023 as fee layers accumulated, and new seller launches hit a decade low in 2025.