How Much Does Amazon Take From Sellers? Where Your Money Goes.

Let me start with a number that should bother you.

$300,000 in sales. $2,800 in your pocket.

That’s not some worst-case horror story I’m dragging out to scare you. It’s the median. Real revenue, real product, real customers buying the thing, and at the end of the month there’s basically nothing left because of the amazon seller fees. These aren’t failing businesses. The founders are working themselves into the ground and taking home less than what a school district pays an aide.

I know how that math happens because I lived a version of it. Years back, before any of this was a company, I was selling on Amazon nights and weekends with a buddy who’d already built a brand and sold it. We figured out the selling part quick. Got good at it, honestly. The part that took us way longer was everything underneath: sourcing, margins, what the unit actually cost us by the time it landed and shipped. We could move product all day. Keeping the money was a totally different skill, and nobody had taught us that one.

So here’s the thing I wish someone had said to me back then. Being good at selling on Amazon is not the same as running a business that keeps money. Most founders can’t say how much Amazon takes from sellers like them once every fee is stacked up, and the blind stop is where that money quietly goes. It’s a dozen small cuts, most of them designed by Amazon on purpose. Let’s walk the money down from the top and find them.

Start at the top, then watch it shrink

Say you do $25,000 in a month. Feels great. That’s the number you tell people at dinner.

Now watch it get smaller. Amazon’s referral fee takes roughly 15% right off the top before you’ve done anything. Then FBA picks the order, packs it, ships it, and bills you for all three. Storage fees, which creep up the longer your stuff sits in their warehouse, and they sit longer than you think. Cost of goods. Freight, which is up for pretty much everybody right now. And PPC, which we’ll get to, because PPC deserves its own paragraph.

Add it all up and that proud $25,000 is now a couple thousand bucks of real margin. And you still haven’t paid yourself, or your accountant, or the four software subscriptions you forgot you’re on.

No single one of these is a scandal. That’s the trap, actually. Each line item looks reasonable in isolation, so you never flag any of them, and meanwhile the stack of them together is the reason you’re putting in sixty-hour weeks for money that doesn’t make sense. The fees were never hiding from you. You just never sat down and put them on one page where you’d have to look at the total.

The PPC leak nobody wants to look at

Here’s the one that actually hurts, because unlike the fees, this one’s on you. You can fix it this week and most people won’t.

A huge number of small brands are losing money on ads without realizing it. The reason is they’re staring at ACOS campaign by campaign, where everything looks survivable, instead of pulling TACOS across the whole account, where the truth lives. Your branded campaigns look fantastic. Of course they do, people searching your name were going to buy anyway. It’s that broad-match top-of-search campaign running at 70-something percent ACOS that’s quietly holding the whole account underwater while you admire the branded numbers.

I’m not saying turn the ads off. Ads are how you win on this platform, full stop. What I’m saying is that “running ads” and “running the ads that actually make money” are two completely different businesses, and most sellers are running the first one while they tell themselves they’re running the second.

So go pull your TACOS for the last 90 days. Total ad spend over total revenue, the whole account, no cherry-picking. If that number makes your stomach drop, congratulations, you just found a leak.

Now the part Amazon just made worse

If the fee math already felt tight, brace yourself, because the cash timing just got worse.

Amazon’s new DD+7 policy holds a chunk of your money for a stretch before it ever hits your bank. Long lead times or a slower carrier? You could be staring at three or four weeks between the sale and the moment you actually see the cash. Sellers are calling it a working-capital extraction strategy dressed up as policy alignment, and I’m not going to pretend that’s an unfair read.

But the annoyance isn’t the real problem. The real problem is that most small CPG brands are running on fumes already, somewhere between zero and fourteen days of cash. So when Amazon stretches the timing out and your co-packer wire is due Friday, you hit the exact moment that takes down healthy companies: profitable on paper, broke in the bank account, same week.

Write this somewhere you’ll see it. Profit and cash are not the same thing. You can be genuinely profitable and still go under because the money showed up two weeks too late to cover the thing that was due. That’s how good brands die, and they almost never see it coming.

Three things to actually do this week

This isn’t a course and I’m not building to a pitch. These are just the first three things I’d do if I woke up tomorrow and this were my brand.

Build the real P&L. Not the Seller Central dashboard, that thing lies by omission. The actual full one. Every fee, COGS, freight, ad spend, software, your accountant, and a real number for your own time. All of it on a single page. I’d bet most founders reading this have never once seen their true unit economics in one place, and you can’t fix a number you’ve never been brave enough to add up.

Split your good ad spend from your bad ad spend. Go find the campaigns running over your break-even ACOS and cut or cap them. Today, not next quarter. Leave the branded stuff alone and protect the non-branded campaigns that are actually converting. This is the rare fix that shows up in days instead of months.

Figure out your real cash runway. If sales stopped tomorrow and Amazon sat on a reserve, how many days could you actually survive? Count it. If the answer is under thirty, then growth is not your job right now. Building a cushion is. Because here’s the part nobody tells you at the conferences: brands almost never die from a bad product. They die from running out of cash on the one Friday it mattered.

The reframe

For years the only conversation in this space was scale. Hit seven figures. Pour gas on it. The aggregators were lining up to buy you out and you’d ride off into the sunset with a check.

That story’s dead. The aggregator dream fell apart, the fees never stopped climbing, and a lot of people who chased pure top-line growth learned the expensive way that revenue you can’t keep isn’t worth much when the bill comes due.

I never built brands on that playbook, and I’m definitely not starting now. The ones that make it from here won’t be the ones with the gaudiest GMV. They’ll be the operators who know their numbers cold, defend their margin, watch their cash like a hawk, and aren’t sitting one policy email away from chaos.

And if you’re in CPG, food, supplements, beauty, anything a customer uses up and reorders, you’ve got an edge a lot of products will never have. People come back. A consumable with a real repeat-purchase rate and a Subscribe & Save base is worth more than a one-and-done product pulling identical revenue, because that second order barely costs you a dime in ad spend. That’s not a growth hack, it’s the most durable margin you can build on Amazon, and if it were my brand it’s the first thing I’d protect and the last thing I’d let slip.

You don’t need to scale faster. You need to keep more of what’s already coming in. Start with the P&L. Everything else gets easier once the truth is sitting on one page in front of you.