Amazon FBA Fees: The Ones You're Overpaying Without Knowing

A client told me about this one. She'd had a previous agency who checked her product dimensions every few months as a matter of routine, and she wanted to know whether we did the same. We did, but I want to be honest about where the idea came from, because it's the single cheapest thing on this list and most brands have never heard of it.

Amazon measures your product itself, bills you on what it recorded, and on flexible packaging it's often measuring the wrong thing. Her flat-packed bags were being recorded as though the gusset were fully extended, which inflated the dimensions and pushed the SKU into a higher fee tier.

Nobody at Amazon was doing anything wrong. A machine measured a bag that can be one of two shapes and picked one.

What does Amazon charge you for FBA?

Before the audit, the map. There are more lines here than most brands carry in their heads.

The referral fee is a percentage of the sale price, set by category and sometimes varying by price point within a category. Pull your own rate rather than working from a number you read somewhere, because this is exactly the kind of detail Amazon revises.

Two details on the referral fee catch people out. Amazon charges it on the total price the shopper pays, and that total includes any shipping or gift-wrap charge. And some categories carry a per-item minimum. On a cheap SKU, that minimum can be what you pay.

The fulfillment fee is per unit, priced on size tier and weight. This is the one this article is about.

Storage fees are monthly and charged on the volume your inventory occupies, with higher rates in the last quarter of the year. Slow-moving stock picks up long-term storage and aged inventory surcharges on top.

Then there's a set of fees that turn on your behavior rather than your product: inbound placement, low inventory, returns processing in some categories, removal and disposal. Each is small. Together, on a catalog with any operational messiness in it, they aren't.

Every one of those is visible in your reports. The trouble is they're visible one at a time, in different places, which is why so few brands can say what the channel keeps in total.

It doesn't help that none of it arrives as a bill. Amazon takes what you owe out of your proceeds before it pays you, pays out every couple of weeks, and holds some money back in case of returns. Storage, returns processing and the fine for a mislabeled shipment come off the same way. Operations leads ask us which of these will land on their desk. All of them do, from the brand's own Seller Central, with nothing marked up and nothing passed through us. If a reseller holds your listing today, you don't see these deductions at all, and that's part of what a reseller keeps when they own your Amazon.

What changed in FBA fees in 2026?

Two dated changes matter this year: a 3.5% fuel and logistics surcharge on FBA fulfillment fees in the US from April 17, 2026, and the holiday peak fulfillment fee, which runs again from October 15, 2026 to January 14, 2027. Both sit on top of the base fulfillment fee. A SKU you priced last spring costs more per unit now, and nothing about the product changed.

Over the last year and a half Amazon has added an enormous number of new ways to nickel and dime sellers, most of them inside the inventory system. That's the reason this article won't print a rate table. Rates published in January are out of date by April, and a margin built on last year's schedule overstates what the channel returns.

Pull the current fulfillment fee page in Seller Central and run your top SKUs through Amazon's Revenue Calculator before you set a price or a promotion, and do it again whenever Amazon announces the next round.

Why the measurement is the fee

Fulfillment pricing works in tiers, and tiers are steps rather than slopes. Cross a threshold on the longest side or on weight and the price jumps, then stays flat until the next one.

That's what makes a measurement error expensive. If Amazon has your product a fraction of an inch over a threshold, you don't pay a fraction more. You pay the next tier up, on every unit, every order, for as long as the recorded dimensions stand.

Flexible packaging is the obvious risk because it has more than one shape. A stand-up pouch flat and a stand-up pouch full are different objects. So is anything compressible, anything with a gusset, anything that ships flat and sits upright on a shelf. Rigid products are safer, though I've seen those recorded wrong too, usually when the original listing was set up by somebody guessing at dimensions and Amazon never corrected it.

Weight has the same shape of problem. Amazon uses the greater of unit weight and dimensional weight on some products, so an item that's light and bulky gets billed on the space it takes rather than what it weighs.

How do I check my own FBA fees?

Half an hour with a tape measure and a kitchen scale, per SKU. It's unglamorous and it's the highest-return half hour in this article.

Start by pulling what Amazon believes. The FBA fee preview report and the fee estimates in Manage Inventory both show the dimensions and weight Amazon has recorded, along with the size tier it has assigned. Write those down.

Then measure the product as it ships. Package it the way a customer receives it, measure the longest side, the median side and the shortest, and weigh it. Compare.

If you'd rather check one product than read a report, Amazon's own Revenue Calculator does the same subtraction for a single ASIN. Search for Amazon's profitability calculator, enter the dimensions and weight you measured, and compare the fee it returns with what you're being charged. Once we have a brand's product dimensions, the fee side of the math is the easy part, and on any new account one of the first permissions we ask for is Manage Inventory. That lets us check the recorded dimensions against the product ourselves.

You're looking for two things. An outright mismatch, where the recorded dimensions don't resemble the product. And a near-miss, where Amazon's number sits just over a tier threshold while your own sits just under. The second is more common and more valuable, because it's the one nobody notices.

Do this on your highest-volume SKUs first. A tier error on a product doing a hundred units a month is an annoyance. On your best seller it's a line item.

How do I request a remeasure?

Through Seller Central support, as a case against the specific ASIN. Amazon will re-measure the unit in its facility and update the record if it finds a discrepancy.

Seller Central also has a self-serve remeasure request. Type "remeasure" into the Seller Central search bar to find it, and fall back to a support case if it doesn't settle the tier.

Two things make the request go better. Send your own measurements, with the product packaged as shipped, so there's something concrete to check against. And do it when you have units in stock at a fulfillment center, because they need something to measure.

If the remeasure lands in your favor, the tier changes going forward. Whether anything comes back for the period you were overcharged depends on the case and on what Amazon's own audit finds, so treat any reimbursement as a bonus rather than the reason to file.

Then diarize it. Packaging changes. Suppliers change film. Someone adds a resealable strip. The measurement that was right last year isn't guaranteed to be right now, which is why the agency in that story was checking routinely rather than once.

How do you avoid storage and low-inventory fees?

Keep each SKU inside a band, because Amazon charges you for holding too much inventory and also for running too thin. My usual guidance is at least 60 days of stock in Amazon and no more than six months.

Above the band you pay monthly storage, which costs more in the last quarter of the year, and then the aged inventory surcharge on whatever sits too long. Below it you start paying low-inventory fees, and if a SKU runs out you lose sales and ranking on top. The same band is what healthy inventory looks like anyway.

Holding the band is a forecasting job. Amazon will suggest how many units to send in, and we don't use its numbers. We watch how each ASIN is selling and send our own, because an automated recommendation doesn't know that one of our brands can only run production once a quarter. Build in the lead time as well. From a brand's dock to sellable at Amazon has run about two weeks for the brands we manage, and longer around Prime Day.

Two inbound decisions add fees that rarely get looked at:

  • Where the shipment goes. When you create a shipment you can send everything to one fulfillment center or split it across several, and the two options carry different fees. Which is cheaper changes from shipment to shipment. Compare them each time instead of setting a default.
  • How the cases arrive. Product goes in as master cases, and Amazon will open a case and sell the units inside as singles unless the pack is sealed and labeled as a set. It won't put them back together. A two-pack has to arrive as a two-pack, or the pack size your margin was built on never reaches the shelf.

How much does Amazon take from a $100 sale?

There isn't one answer, because the referral fee moves with your category and the fulfillment fee moves with size and weight. The only honest figure is the one you get by running your own SKU through the subtraction.

The two big fees behave differently as price rises. The referral fee is a share of the sale: on a $100 order, it's your category rate in dollars. The fulfillment fee is set per unit by size tier and weight, and a product costs the same to fulfill whether you sell it for $15 or $100. The cheaper and heavier the product, the larger the share fulfillment takes, and on a low-priced product it can approach 30% of net sales.

You'll see a rule of thumb on seller forums that Amazon keeps about a third of the price before advertising. Treat it as a reason to check your numbers rather than a number to plan on. When we first look at a prospect's margin from the outside, it's almost always missing their own shipping, returns and the smaller charges Amazon adds to the P&L, and it comes down once those go in.

What does the whole fee stack look like?

Here's why I care about this more than the size of any one fee.

We built a client a year-to-date Amazon P&L: net sales at the top, every fee subtracted line by line, down to real net operating profit. All-in Amazon fees ran a little over 40% of net sales. Fulfillment alone was close to 30%. Advertising was just under 15%. What survived to net operating profit was around 20%.

That advertising line is TACoS by another name, and what a good TACoS looks like depends on whether the spend is finding new customers or charging you for ones you already had.

The brand had never seen that page. Not because anyone hid it, but because Amazon doesn't produce it. Amazon knows what it sold and what it charged you. It has never seen your manufacturing invoice or your freight bill, so it can't do the subtraction, and no dashboard inside Seller Central will do it for you.

When fulfillment is close to 30% of net sales, a single fee tier is not a rounding error. It's a share of the largest cost line in the business, and it's the one cost line you can move without changing anything about the product, the price or the advertising.

Can a different pack size lower your FBA fees?

Often, because the fulfillment fee is charged per unit shipped, so a multipack spreads one fee across more product and raises the order value the fee is measured against. It's one of the first levers I reach for on a SKU under the margin floor, before anyone talks about cutting cost.

Your retail pack was designed for a shelf and a planogram. On Amazon it's usually the wrong size, and a single small unit is where fulfillment takes its biggest bite. A three-pack or six-pack of the same product pays one fulfillment fee, usually at a higher tier, where three or six singles would each pay their own, and an Amazon-only pack also keeps your listing out of the per-unit price comparison that can cost you the Buy Box. We cover the margin side of this, and how to keep an Amazon pack from undercutting retail, in the piece on the 30% pre-ad margin floor.

Bundles do the same work for products that aren't sold in multiples. The easiest place to find one is your own website. Look at which products customers already put in the cart together, build that pairing as a set, and test it.

Is FBA still worth it in 2026?

For most of the brands we work with, yes. Nine times out of ten we recommend FBA, and the exceptions are frozen, refrigerated and meltable products, plus some heavy ones where fulfillment eats too much of the sale.

The test is per SKU and it happens before advertising: sale price, minus true landed COGS, minus every Amazon fee. If that clears 30% on a consumable, FBA's fees are affordable, and for most consumables the Prime badge they buy earns its keep. Under 20%, we won't take the account on until the pricing or pack size changes, because no amount of ad performance makes up the gap.

With fees up this year, some SKUs that passed last spring won't pass now. Put the audit on the calendar.

When fulfillment is the biggest line, question FBA itself

If you run this and find fulfillment eating close to a third of the sale price, the tier audit is the first move and the second is bigger: ask whether FBA is right for that SKU at all. Merchant fulfillment can be dramatically cheaper on heavy or bulky products, and it carries traps of its own around handling time and review liability. We've worked through that comparison in the FBA versus FBM piece.

The answer is per SKU, like everything else here. Heavy and slow-moving is where FBM usually wins. Small, light and fast-moving is where FBA usually does.

The margin that goes to zero

One more thing about fees, and it's the one that gets left out of every article like this.

A founder I spoke to runs a perishable line at margins around 60%, which sounds like a product with nothing to worry about. Then a box arrives melted. The order refunds, the unit is unsellable, the return is processed, and that transaction is at zero or a net loss. Two of those a month against a healthy average and the average stops describing the business.

Fees and returns are the same conversation because both are variance, and average margin is a poor planning number when one failure mode wipes out an order entirely. If you sell anything heat-sensitive, fragile or perishable, model the failure rate as a cost line rather than treating it as an exception.

There's a visible penalty attached too. Amazon labels products with high return rates in some categories, and that badge sits on the listing where shoppers see it. Returns are a fee, a lost sale and a conversion problem at the same time.

On a young catalog, one or two returns move the return rate several points either way. Read a new listing's return rate over a few months before deciding the product has a problem.

The best diagnostic is one any brand selling in more than one place can run. If a product gets returned on Amazon and almost never comes back through your own website, the Amazon photography or description is probably setting an expectation the product doesn't meet. If your Amazon return rate is well above your direct one, you're paying for a listing problem as a fee, and the cheapest fix is on the page.

What to do this month

Pull the FBA fee preview for your top ten SKUs by volume. Measure and weigh those ten products as they ship. File remeasure cases on any that don't match, sending your own numbers with the case. While you're in there, check how many days of stock each of those ten has in Amazon, and flag anything under 60 days or over six months.

Then build the full page. Net sales, every fee subtracted, true landed COGS, advertising, down to what the channel returned. Per SKU where you can hold it, because a blended catalog average hides the loser inside the winner.

Most brands find something in the first pass. Sometimes it's a tier. More often it's the total, which nobody had put on one page before.

The bottom line

Fees are the part of Amazon that feels fixed, which is why almost nobody audits them. Most of it is fixed. The measurement isn't, and it sets the largest per-unit cost in the business.

Half an hour with a tape measure is a strange thing to recommend to a brand doing serious volume. Do it anyway, on your top ten, and then build the page that shows what the channel returns.

If you'd rather see that page built properly, we'll do it. Send your COGS and product dimensions and we'll build your Amazon P&L: net sales down through every fee to real net operating profit, per SKU, whether or not there's a conversation after it.

cta_free_trial

Read more blogs

How to Increase Amazon Sales Without Buying Them: The Amazon Profit System
  • Amazon

How to Increase Amazon Sales Without Buying Them: The Amazon Profit System

How to Increase Amazon Sales Without Buying Them: The Amazon Profit System
Flat for Two Years, Then +60%: A Specialty Ingredient Brand's Amazon Turnaround
  • Amazon
  • Case Study
  • Natural CPG

Flat for Two Years, Then +60%: A Specialty Ingredient Brand's Amazon Turnaround

Flat for Two Years, Then +60%: A Specialty Ingredient Brand's Amazon Turnaround
From Break-Even Ads to Roughly 10x Profit: A Natural Snack Brand's Amazon Turnaround
  • Case Study
  • Natural CPG
  • Amazon

From Break-Even Ads to Roughly 10x Profit: A Natural Snack Brand's Amazon Turnaround

From Break-Even Ads to Roughly 10x Profit: A Natural Snack Brand's Amazon Turnaround

Find out what your Amazon channel really earns

Apply for a free trial of our work, or chat with our team. Either way you'll leave knowing your true profit, which is more than your current reports can tell you. And if we work together and miss the targets we set, the sprint to fix it is on us.