Amazon's all-in take rate in 2026, line by line

Every operator who sells on Amazon knows the 15% referral fee. Almost none of them build their P&L on it, because 15% is the starting bid, not the settlement. By the time Amazon has taken its fulfillment fee, the new fuel surcharge, inbound placement, storage, and the ad spend you now need just to be visible, the real deduction on a typical consumable is closer to 39% of revenue. That is before you have paid a cent to manufacture the thing. This post runs the exact math on a $40 SKU using the 2026 fee schedule, names every line, and shows which price points and categories can actually carry the full stack.

The published rate is the starting bid, not the story

Amazon's referral fee is the number every seller can recite: 15% in most categories. It is also the most misleading number in the whole model, because it is the one fee that stays constant while five others pile on behind it.

Independent analysis of seller profit-and-loss statements has been flagging this gap for years. Marketplace Pulse, which builds its estimates from a sample of real seller P&Ls, found the effective take rate crossed 50% back in 2022, up from around 40% five years earlier. Its March 2024 follow-up put a typical private-label seller at 50 to 60% of sales going to Amazon in fees once advertising is counted. Bloomberg and the American Prospect ran the same finding: Amazon's cut of each third-party sale has roughly tripled over a decade.

When I talk to founders running a brand in the $5M to $20M range, the pattern is almost always the same. They modeled the channel on the referral fee, launched, and then discovered the real number only when the first settlement report landed. The fees come out before the deposit hits, so the P&L they built in a spreadsheet and the cash that shows up in the bank are two different stories. The goal of this piece is to make sure you see the second story before you commit inventory.

One caveat on that take-rate line: those figures include advertising as a seller-side cost and come from a sample of P&Ls, not from Amazon's own books. The direct-fees-only number, before ads, is lower, roughly 24 to 30% for a standard $40 SKU. Both numbers matter. We will build up to them line by line.

The 2026 fee stack, line by line, on a $40 consumable

Take a Health and Personal Care consumable: $40 selling price, large standard size, about 12 ounces, sold through FBA. Here is what Amazon deducts, in order.

Referral fee. 15% of the selling price, so $6.00. Flat, no tier, no way to reduce it short of changing category. This is the one fee you cannot engineer away.

FBA fulfillment fee. For a large standard-size item in the 8 to 12 ounce band, the 2026 base fee runs about $4.10. Add the 3.5% fuel and logistics surcharge that took effect April 17, 2026, and it lands near $4.24 per unit. This is a fixed dollar amount, which matters enormously for cheaper SKUs, as we will see.

Inbound placement fee. Roughly $0.35 per unit for a standard-size item in a typical minimal-split shipment. You can drive this toward zero by spreading inventory across four or more fulfillment centers, but most brands shipping consolidated pallets pay somewhere in the $0.21 to $0.68 range.

Storage. For a 0.3 cubic foot box held on a 30-day turn during the off-peak season, storage is about $0.23 per unit at the $0.78 per cubic foot rate. Cheap here, but this is the line that detonates in Q4, covered below.

Advertising. This is the one operators still treat as optional and should not. At a 12% TACoS, the low end of the established-brand range, ad spend is $4.80 on a $40 sale. In Health and Personal Care, where cost-per-click runs north of $1.40, launch-mode brands routinely run 18 to 25%.

Fee line	Basis	Dollar amount	% of $40
Referral fee	15% of selling price	$6.00	15.0%
FBA fulfillment (base, large std 8 to 12 oz)	2026 schedule	$4.10	10.3%
3.5% fuel surcharge (since Apr 17, 2026)	3.5% of fulfillment	$0.14	0.4%
Inbound placement (minimal split)	2026 schedule	$0.35	0.9%
Storage, off-peak (30-day turn, 0.3 cu ft)	$0.78/cu ft	$0.23	0.6%
Total direct fees (before ads)		$10.82	27.1%
Advertising, 12% TACoS	12% of revenue	$4.80	12.0%
Total all-in Amazon deduction		$15.62	39.1%
Brand keeps before COGS		$24.38	60.9%

Now bring in your own product cost. If landed COGS is $12 (30% of price), contribution margin is ($24.38 minus $12.00) divided by $40, or about 31%. Workable. At $16 COGS (40% of price), you are at 21%, viable but thin. At $20 COGS, contribution drops to 11%, and a single return spike or a Q4 storage overhang tips the SKU into a loss.

Why price point is the lever, not the category

The fixed FBA fee as a percentage of revenue is the whole game. That $4.24 is 10.6% of a $40 product but 28% of a $15 one. The referral fee scales with price, so it is neutral, but the fulfillment fee, placement, and storage are all roughly fixed per unit. That means the cheaper your SKU, the larger a bite Amazon takes as a percentage.

Selling price	Referral	FBA fee*	Inbound + storage	Direct fees	Direct %	+ 12% ads	All-in %
$15	$2.25	$4.24	$0.58	$7.07	47.1%	$1.80	59.1%
$20	$3.00	$4.24	$0.58	$7.82	39.1%	$2.40	51.1%
$30	$4.50	$4.24	$0.58	$9.32	31.1%	$3.60	43.1%
$40	$6.00	$4.24	$0.58	$10.82	27.1%	$4.80	39.1%
$50	$7.50	$4.24	$0.58	$12.32	24.6%	$6.00	36.6%

Below about $25, the all-in take passes 46% and the room left for COGS plus profit gets uncomfortably thin for a normal CPG cost structure. This is why so many sub-$20 Amazon products are sold as multipacks: bundling three units into a single $35 to $45 order spreads that one fixed fulfillment fee across the pack instead of paying it three times.

Category matters too, but less than price. Referral fees range from 8% (Grocery under $15, Consumer Electronics) to 15% (most consumables, Home and Kitchen, Sports) up to 17% for apparel over $20 and a punishing 45% for Amazon Device Accessories. For a sense of how other marketplaces stack up, the Etsy teardown on take rate and growth shows a different fee architecture with similar margin pressure. If you sell food or beverage, the Grocery sub-$15 carve-out at 8% is a genuine escape hatch worth designing your price points around.

The margin structure decides the whole thing. When I talk to supplement founders, the ones running 75% gross margins going into Amazon can absorb even a 25% TACoS and still hold contribution, because the fee stack is being taken out of a fat margin. A lower-margin consumable at 45% gross simply cannot carry the same fees. Same fee schedule, completely different outcome, and the difference is entirely on your side of the ledger.

The Q4 storage trap and the aged-inventory surcharge

Storage looks trivial in the base table at $0.23 per unit. Then Q4 arrives. Standard-size storage jumps from $0.78 per cubic foot off-peak to $2.40 from October through December, a 3x surge. On that same 0.3 cubic foot consumable, monthly storage goes from about $0.23 to $0.72 per unit. On a slow-moving SKU that $0.49 per unit jump can erase a full month of contribution margin all by itself.

There is a second, newer trap. Effective January 16, 2026, Amazon added an aged-inventory surcharge tier for stock sitting 456 days or longer: $7.90 per cubic foot or $0.35 per unit, whichever is greater, stacked on top of normal storage. On our 0.3 cubic foot sample box, the per-cubic-foot rate governs: 0.3 times $7.90 comes to about $2.37 per unit per month. The $0.35 is a per-unit floor that protects Amazon on very small items; most standard-size consumables pay the cubic-foot rate. For a brand that over-ordered and is sitting on dead stock, that is a compounding penalty for capital you have already tied up.

The pattern we see again and again is brands that treat Amazon like a warehouse they can stuff for the holidays. It is not. Sending your entire Q4 buy in September to "be safe" means paying peak storage on inventory that will not sell until December. The move is to ship in waves timed to sell-through, keep 30 to 45 days of cover in the network, and hold the rest at your 3PL or with your manufacturer where storage is a fraction of the cost.

What this means for your P&L before you launch

Here is the operator rule that falls out of all of it. If your gross margin before Amazon fees is below 40%, the fee stack will most likely leave you loss-making on FBA, and no amount of ad optimization fixes a structural margin problem. Between 40 and 60% gross margin, the channel works if you keep TACoS disciplined and your SKUs above roughly $25. Above 60% gross margin, Amazon is usually a strong channel even with heavy ad spend, which is why supplements, skincare, and other high-margin consumables dominate the platform.

Before you commit inventory, stress-test four inputs, because these are the ones that move the answer: COGS as a percentage of price, your realistic TACoS (not the aspirational one), your return rate, and your seasonal storage exposure. Model the bad case on all four at once, not one at a time. A SKU that pencils at 12% TACoS and a 3% return rate can flip negative at 20% TACoS and 8% returns, and both of those can happen in the same quarter.

The practical inverse is worth memorizing. Your minimum viable Amazon price is roughly (COGS + fixed Amazon costs) divided by (1 minus your target contribution margin minus the variable Amazon fee percentages). Run that before you set a price, not after, and you price the channel instead of letting the channel price you. If you want a second set of eyes on the model before you commit a purchase order, that is exactly the kind of question our fractional CFO services for ecommerce brands exist to pressure-test.

The 15% referral fee is the number Amazon wants you to plan around. The 39% all-in deduction is the one that actually hits your bank account, and on cheap SKUs it is closer to 55 to 59%. Model the second number before you ship a single pallet, or the settlement report will do the modeling for you.

Sources and methodology

The worked example uses Amazon's official 2026 fee schedules. Referral fees are from Amazon Seller Central reference G200336920; FBA fulfillment fees from GABBX6GZPA8MSZGW, effective January 15, 2026; inbound placement from GC3Q44PBK8BXQW3Z; and monthly storage from G3EDYEF6KUCFQTNM. The $40 case assumes Health and Personal Care (15% referral), large standard size (~12 oz), a minimal inbound split, a 30-day off-peak turn, and 12% TACoS. These are modal assumptions, not best or worst case.

The 3.5% fuel and logistics surcharge took effect April 17, 2026. It applies on top of FBA fulfillment fees. Coverage of the change and its context is in Supply Chain Dive. On a ~$4.10 base fee it adds roughly $0.14 to $0.16 per unit. Sellers should confirm the exact fee for a representative ASIN in Amazon's own FBA revenue calculator, since fees vary by the precise weight and dimensions.

The effective take-rate figures are independent estimates, not Amazon disclosures. Marketplace Pulse builds them from a sample of seller P&Ls in Amazon Fees Only Go Up (March 2024), which puts a typical private-label seller at 50 to 60% of sales in fees including advertising. The decade-long trajectory from 19% (2014) to the mid-40s and beyond is documented in the American Prospect's Amazon's Latest Seller Squeeze (January 2025). These figures include ad spend as a seller-side cost; the direct-fees-only number is lower.

Advertising benchmarks come from category ad-performance data. TACoS of 10 to 15% for established brands and the Health and Personal Care cost-per-click range are drawn from published Amazon advertising benchmark reports for 2026. A brand in launch mode or in a high-competition category will run materially higher, which is why the model treats 12% as the floor of a realistic range, not a target.

What the base calculation excludes. Return processing fees, low-inventory-level fees, removal fees, and the monthly Professional plan fee are conditional and left out of the headline stack. Return processing ($2.12 to $7.90 per unit) is immaterial for low-return consumables but material for apparel and higher-return categories. Treat the 39% figure as the clean-case floor; real settlements usually run higher.

Frequently asked questions

what percentage does amazon actually take from my sales in 2026?

For a $40 consumable using FBA, the direct fees (referral, fulfillment, placement, storage) run about 27% of revenue. Add a normal 12% ad spend and you are near 39% before your own product cost. Independent seller-P&L studies put the typical private-label seller at 50 to 60% once heavier ad spend and returns are counted.

what is the minimum price point where amazon fba makes sense?

For a standard-size consumable, roughly $25. The FBA fulfillment fee is a fixed dollar amount, so at $15 it eats 28% of revenue and the all-in take passes 59%. At $25 and up, that fixed fee shrinks as a share of price and you keep enough room for COGS plus a real margin.

what's the difference between acos and tacos and which should i use to model amazon profitability?

ACOS is ad spend divided by ad-attributed sales, so it only measures the paid slice. TACoS is ad spend divided by total sales, paid and organic. Model your P&L on TACoS, because it captures ad spend as a share of everything you sell. ACOS is for tuning individual campaigns.

how much does the amazon fuel surcharge add to my fba fees?

Amazon added a 3.5% fuel and logistics surcharge on FBA fulfillment fees effective April 17, 2026. On a ~$4.10 base fee that is about $0.14 to $0.16 per unit. It is small per unit but it stacks on top of the January fee increases, so year over year the same SKU costs more to fulfill.

what happens to my margins during q4 when storage fees triple?

Standard-size storage jumps from $0.78 per cubic foot off-peak to $2.40 in October through December. On a 0.3 cubic foot consumable that is roughly $0.23 versus $0.72 per unit per month. If your SKU turns slowly, sending heavy Q4 stock can wipe out a full month of contribution margin on that product.

is it even worth selling on amazon with all these fees?

For consumables and repeat-purchase products, usually yes, because that is where buyers already search and reorder. The channel works well when your gross margin before Amazon fees is 60% or higher. In the 40 to 60% gross margin band, the channel can work if you keep TACoS disciplined and your SKUs above roughly $25. Below about 40% gross margin the fee stack leaves you loss-making, and you are better off pushing those SKUs through your own store.

how do amazon fees compare to selling on my own shopify store?

On Shopify you pay roughly 2.9% plus $0.30 payment processing plus your own fulfillment and ad costs, so the platform take is far lower. The tradeoff is you have to buy the traffic Amazon hands you for the referral fee. Most brands run both: Amazon for discovery and reorders, their own store for margin and customer data.

does amazon charge more than 15% referral fee in any category?

Yes. Amazon Device Accessories is 45%. Clothing above $20 is 17%. Most other categories sit at 8 to 15%, and some, like Grocery under $15 and Consumer Electronics, are 8%. Check the exact category before you model, because the referral line is the one fee you cannot engineer away.