78 comments

[ 2.7 ms ] story [ 142 ms ] thread
The current headline on both the article and HN submission ("Amazon Takes a 50% Cut of Seller's Revenue") is not accurate; the actual base rate is 8-15% according to the article. If a seller elects to use Fulfillment by Amazon, that adds another 20-35%, and advertising on Amazon can add up to another 15%, again according to the article.
FBA prices are ok-ish. You can undercut them by using your own ligostics. But to do so you have to know what you do logistics wise, and most sellers don't. Multiplatform selling was one of the main reasons to quit FBA, but Amazon started, if memory serves well, a service that covers sales over different marketplaces. Plus, FBA gets you Amazon Prime and that coveted Buy Box, doing the same as a marketplace seller is possible but hard.
>Plus, FBA gets you Amazon Prime and that coveted Buy Box, doing the same as a marketplace seller is possible but hard.

Correct.

My FBA and FBM margins are very comparable. What is not comparable is the colossal boost in sales that comes from getting the buy box and Prime logo.

Walmart takes a different approach, and as a result its FBA equivalent isn't nearly a sales booster. <https://np.reddit.com/r/WalmartSellers/comments/xh2zd5/walma...>

The 50% figure is accurate, though. It's written in the most shocking way possible, but it's not incorrect. I'm not saying that I agree that 50% is overpriced, since electing to do your own fulfillment and advertising elsewhere will still represent an expense. Sellers may very well be pay just as much or more if they take care of their own fulfillment and advertise elsewhere. I don't know.
FBA is not a fuction of the unit price though, you say 20-35% which may just be what it ends up in, but its based upon package dimensions and weight.
Ah, nice to see someone with insights to the inside:-)
They don't "take." They offer a price for selling on their site, and sellers choose to pay it or not. They built that better mousetrap.
If they have a monopoly control of the market, then the prices could be higher with sellers having little choice between paying and not selling at all.
I think you might be getting at Monopsony, which is the only place for sellers to sell to. Monopoly means the only place to buy from.
Well the sellers aren't selling to Amazon.

Monopoly is the most generic term so I don't think it's 'wrong', and technically neither are correct so...

Monopoly also means "exclusive or near exclusive access or control of something" for instance "she is monopolizing her time" and in that usage the original poster is using it correctly.

In fact, I think we should retire the term monopsony exactly because its actual usage is a subset of the usage of monopoly and nobody ever cares about the distinction.

(comment deleted)
Does Amazon have a monopoly on the market, then? Because if not, I don't see your point
It depends on the exact market segment (Amazon does not have a monopoly on groceries, for instance), but yes, I think Amazon is effectively a monopoly. Or at least part of an oligopoly, which is no better.
I don't see how you can think this. Amazon is not the only payment processor, it's not the only inventory management company, not the only delivery company, not the only online store company.

There are any number of permutations of different service providers you could use to sell online. In fact, a very large number of people who sell online don't use Amazon at all. Or they will sell on Amazon as well as in other channels, or even direct.

For mainstream retail in the US, you have Amazon, Walmart, and... who?
Target, Nordstrom, etc?
Yes, I think you could argue Target. But Nordstrom? I don't think they have that sort of power.

Remember, an oligopoly (or even monopoly) doesn't necessarily mean there aren't other players, it means that the oligopoly is the only thing that can really influence the market. It's about who is writing "the rules", so to speak.

Amazon has a monopoly on creating a web page and selling and shipping stuff? There are literally thousands of merchants that setup their own sites or sell through Facebook, Instagram or Etsy
I don't know if I would characterize them as having monopoly control of the market because nearly 100% of all of my purchases are made online and 0.0% of my purchases have been made using amazon.com (or walmart.com).

Generally speaking I prefer to buy a product directly, or from a specialized retailer.

Last week I purchased a stethoscope from an online stethoscope retailer. A benefit of doing this is that they offer laser engraving-- something that Amazon, with its warehouses of robots and robot-like employees, cannot easily do.

The specialty retailer also carries a wider array of color and finish combinations, because that's their product niche.

Using a specialist retailer has allowed me to flex on my fellow volunteer EMTs with a personalized Technicolor stethoscope, which is dumb but nice.

You are looking at the wrong side of the market. You need to look at the market from the perspective of the seller v

For examples, do the 7 bigges supermarkets in britain, Tesco, Lidl, etc, have a monopoly?

I can buy all my food from the snaller shops and market markets, as a buyer.

But if I am a seller, and I need to sell mass market food, and all the big supermarkets refuse to carry it, you will go bankrupt.

"Take" can also mean "accept" or "require".
> The 15% transaction fee has stayed the same for over a decade. It varies by category and can be as low as 8%. Fulfillment by Amazon (FBA) fees have steadily increased. Amazon has raised fulfillment fees every year and introduced increases in storage fees. Selling on Amazon is tied to using FBA, so it’s rare for sellers to be successful without using it.

This really is a questionable spin of something that really isn't that complicated. If you sell through Amazon's warehouse, shipping and logistics system they take a larger cut but if you do that you also don't really have to think about how your product reaches the customer. You just ship them your stock and they take care of it.

So in that context I'm not sure I understand why this would be a bad deal. Doing the last mile delivery, sorting and storing stocks is a big part of costs for a seller so while the number might look high that's the cost of doing business.

This is exactly it.

And the increase in total fees is a transfer of marketing budgets from other platforms like Google Ads and Facebook ads to Amazon Ads.

If anyone thinks 50% cut is a big margin they probably haven’t operated an e-commerce store the traditional way.

Apple, Steam, Google et al. charge 30% on *digital* goods. With Amazon you get <50% for physical goods. If anything that's impressive.
"If anything that's impressive."

I'm conflicted about this statement. On one hand, it's almost passive revenue for the seller. On the other hand, Amazon offloads the risk of carrying inventory and takes half the revenue.

I'd argue that it's impressive with one caveat: the "partnership" heavily favors the house.

(comment deleted)
Important to note that plenty of other companies will carry the inventory and ship goods out for you (Amazon did not invent the concept of third party logistics) - however if you go to a third party warehouse it is difficult to get certified as 'Prime'.

(Seller Fulfilled Prime is a thing, but in reality is currently closed for sign-up and is really only for the large sellers)

Apple, Google, maybe others charge 15% on up to $1 million or the first $1 million in digital sales. Or less when negotiated. Does Amazon offer reduced rates to small businesses or low volume sellers?
Digital goods have very low marginal unit costs compared to physical goods.
Edit: I was wrong :)
(comment deleted)
You are misunderstanding the concept of marginal unit costs if you disagree on that basis.

AAA software is wildly expensive, but each incremental copy sold is not wildly expensive.

With a tea kettle, each incremental copy still costs you metal, heating elements and plugs, so the unit margins are smaller.

Oh! That I failed to recall. You are absolutely correct.
It tells me that the fee on digital goods is insanely high.
It is not that I think 50% is a lot but it is the same 50% when amazon is selling fakes. If amazon would be able to sell items that are original and the money would go to the right person maybe 50% is ok. This is not the case by a long shot.
As a former Amazonian who has worked on vat, compliance, taxes, product compliance etc, I can tell you, really from an unbiased perspective that Amazon is doing everything they can to terminate the bad actors.

This includes the fake reviews etc. Lets say you have a charger for iphones, you cant have text for the ASIN saying for iphone, it will get delisted immediatelly, you have to say "compatible with iphone" and such.

The main weakness is how products are created, the catalogue principle, this allows some to fly under the radar for a bit sometimes. And Amazon will not immediatelly shut down a whole sellers account, but rather target ASINs one by one. The best way for a buyer to deal with this is to leave a sellers(not product) review and ask for a refund, you will always get the refund, Amazon always pushes liability to the sellers in order to protect the brand image and to not be exploitable. Once theres enough of reviews and refunds, the sellers will get shut down and the funds will be held on hold, just like a banl does for credit card processors. The cost to do business on amazon with fakes repeatedly(need new companies every time etc) is quite high, if not prohibitivelly high. Do not suck it up, do the 2 things mentioned to keep the place tidy.

Amazon is doing nothing about the total product switches where hard drives have reviews about laundry detergent. They’re clearly not doing basic easy prevention.

Customers try really hard to report stuff like this and Amazon actively thwarts them.

It’s insane to say that Amazon is trying really hard to prevent fraud. Maybe some types of fraud, sure. But so much of the customers-facing fraud is actively supported by Amazon, like Amazon puts actual active effort into preventing customers from reporting it!

This is why I said leave bad reviews on the seller, not the item. You can report brand infringement.

Often, you might need an account or a seller account, this could be improved indeed, however, whats available:

Amazon.com https://www.amazon.com › report Report Infringement

https://www.redpoints.com/blog/report-infringement-amazon/

https://www.redpoints.com/blog/remove-counterfeit-amazon/

Going to customer service will not lead anywhere usually, it needs to end up with the dedicated team.

As a former Amazonian too, I can tell you that we internally had several feature request open to certain teams who could not give two shits about customers anymore. We (a bunch of engineers working on the website platform) we enraged about the incompetence some other teams were showing when it comes to customer satisfaction. And I know for a fact that Amazon is not doing everything they can, not by a long shot. The problem is that the way how Amazon operates the quality is really all over the place, depending on teams much more than in other companies.

I used to buy everything on Amazon and slowly transitioned to traditional stores and only buy things that I cannot in an offline store. Btw. the quality declined over time so you can't say that it was never great. 10 years ago you could have much more confidence that you get what you paid for then today.

Hi there, nice meeting you. Yes, as a dev, it will be hard to push feature requests, I have realized this much, the whole query tools and seller central are rigid and ther are few changes. It is probably very hard to get features approved because there are many stakeholders involved etc.

There is indeed a massive segregation between the teams, the only correspondence is via the ticketing system and sometimes dms via Chime, which happens rather slow, if at all. Now, it depends who you mean by customers, Amazon does not consider the end buyers as first class customers, their customers are the sellers. I know what you mean by some "rogue" teams, without naming any by name, but it is core departments. There exist some parallel departments, like literally doing the same tasks, but by different procedures, as many teams develop their own procedures(for example the vat teams did everything from scratch and even the andon advisors struggle to follow on that). I have seen it happen, teams doing the very same tasks, but in totally different ways, neither is wrong, but it does cause confusion, usually its the older teams not catching up with new developments, from what I have seen. The "SOPs" are more applied on a global level rather than on a team layer.

I am not sure if the US market is different than Europe and Uk, it could be, I never had access to US SC. I do not shop online often, maybe 10 times in 10 years, most of it via Amazon, simply because I get from them what I order and very fast. Any other platform I had tried I regretred very fast.

Or lets ask it this way? How could amazon ensure that all the products are legit before they are being delivered? I am aware of comingling, which can be disabled afaik. But its impossible to check every item before its shipped and if its seller fullfilled orders, there is no way either. I wouldnt know how to improve that other than shutting a sellers account down upon first report, which then leaves you open to false claims by competitors etc. For example, if you want to hurt a competitor, you could just publish many fake 5 star(not 1 star) product reviews in an unreasonably short time(higher than usual frequency) and that will likely be the death knell for the review feature, it would be disabled.

Of course, I have also seen dog collars falsly being flagged as baby toys and the AI findig ridiculous dead end path for case handling and the translation bots having impeccable, even most eloquent language skills, but the conrent being way off. Sometimes these bots get the "needs no human review" approval, when they are not ready for business.

And dont get me started on the "outsourcers":-)

From what I have seen, most dev work is thrown at SC and the various ASIN back offices for the catalogue content. Seen some rather bad db queries for the oracle instances and shared workbook usage where you rather shouldnt make that choice etc. But the website front end gets the least "love".

I should habe qualified my statements with "as far I am aware".

> How could amazon ensure that all the products are legit before they are being delivered?

Better question yet, should Amazon exist in this form if it cannot ensure that all products are legit? We quite often assume that the current for of existence is the only way to go about the problem.

With the power of pki and blockchain it is trivial to create a platform (I know because I was part of a team that created one) where traceability is a feature and it is impossible to game the system the same way it is possible now with the current fulfilment situation.

This is a legitimate question, most likely the hordes of legal and compliance have ensured its "legal".

They offer the refund remedy, but i suspect(actually, I am sure, I have worked in banking and fintech before) that is to prevent chargebacks.

It is surprising how many chargebacks are mitigated this way, if any merchant, even Amazon, gets hit by too many chargebacks, the acquirer will shut them down if heavy penalties are no remedy.

But knowing Amazons order volumes, it is probably still too lucrative for the acquirers to sack Amazon as a merchant.

Amazon should do something about it, you have a good point, they are good at shuffling liability, the concept is one of a kind, but they should apply the same to product integrity and QA.

Someone somewhere is seeing the numbers and has not seen a need to intervene on the matter yet. I am sure you remember the tree charts of teams, all starting and leading to the c level, right?

Its either c level or the c level reports making these decisions.

I think removing comingling and strictly banning the infringing accounts upon first report should solve this issue 99%. Block, hold the funds, review and make a final decision. I would not tolerate mislabeled storage devices or fakes.

For example, there have been aldo shoes for impossible prices on there, I think Aldo choses not to sell on Amazon, so it might take a while before someone realizes that. Usually the system detects such things quite well, if a company has a brand and a patent, they are supper protected on amazon, for example selling boots looking anything like the docs and mertens brand will get shut down fast and efficiently. Maybe some brands are more protected than others, I do not know.

If the seller has to ship the product to Amazon anyway, why not just ship it to the customer instead and not pay the extra fees? What need is Amazon fulfilling that Fedex or DHL do not?
Storing the product and packing it up. If you have small volume, then garage might work and can pack yourself. But Amazon and customers have expectations on speedy shipping. For larger volumes, you need warehouse and people to package orders. Doing yourself makes sense if already have own store but otherwise probably easier and cheaper to pay Amazon.

Also, my impressions is that Amazon shipping is cheaper since they have volume discounts, use their planes for long-haul, and use their trucks for cities.

That makes sense I think. So as a seller, I could buy bulk from some producer, get it bulk shipped to Amazon, and they will take care of meting it out to individual purchasers. Did I get that right?
Yes but competition between sellers in most niches is so tight that as a new seller you'll struggle to make profit, even if you've innovated on the product, because you'll be paying Amazon for search engine ads and also dropping your price as far as you can to win sales. You essentially end up as an unpaid employee of Amazon, managing and optimizing a listing.
Ever ask that of your grocery store? Do you order cereal from Kelloggs.com?
Mostly 1 and at worst 2 day delivery for the entire continental US. You would need between 3-6 warehouses around the US to meet this. Which isn’t a small starting place for most sellers.
> Mostly 1 and at worst 2 day delivery for the entire continental US.

Amazon does not offer reliable shipping speeds and most certainly does not make any shipping speed guarantees. If you want worst-case two-day delivery, you can't use Amazon, because they don't do that.

The near 300 packages a year they deliver to me disagrees.
Assuming you are the only customer for all products sold through Amazon, that's good evidence.
Maybe they get more sales that way?

If I can pick between two items on Amazon for the same price and one is fulfilled by Amazon, I buy that one.

Next day delivery for "free".

Price out overnight service from Fedex or DHL. It's frightfully expensive.

I actually run an FBA business.

For me to ship 500x units to the warehouse, from my home in Australia, it costs around US$200. Factor in the FBA fee of around $3 per unit, and you get a total shipping and logistics cost per batch of $1700.

For me to ship 1x unit from my home in Australia to the US, it would cost around US$20 and take 2 weeks to get there. In this instance, I've paid $10,000 and the customer experience is worse.

Even if I were based in the US, I'd be looking at $6 per package via USPS, plus all the fucking around to label and box up those 500x units. (That's maybe $3200 total including packaging, plus about 10 hours' labour distributed over the course of two months.)

I guess the point is that making use of highly-efficient logistics services allows both small businesses and courier companies to access Amazon's economies of scale.

> For me to ship 500x units to the warehouse, from my home in Australia, it costs around

> For me to ship 1x unit from my home in Australia to the US, it would cost around US$20 and take 2 weeks to get there. In this instance, I've paid $10,000 and the customer experience is worse.

Why are the units starting from your home? Is that where they're manufactured?

Logistics and order destination and they will work 247, give or take.

There is quite some overhead if you have to schedule all the pick ups and update the order and shipping status. They have all this automated and they have warehouses strategically located everywhere.

This is something that many of the chinese sellers still do not use. They are relatively new and inexperienced, they struggle to do the corporate onboarding paperwork etc. They will get around, though, some have improved fast.

The FBA fees are interesting. I briefly talked to someone who runs a store on Amazon. Yes you’re paying Amazon a fee for storage and other logistics, but as a seller get the added benefit of your product just being delivered to the customer with strong delivery guarantees (even fedex and ups give Amazon packages a higher priority over your regular packages). All you have to worry about is get your product to them and they will distribute it over a chain of warehouses by doing their own forecasting of demand to get the optimal delivery experience. Sellers actually don’t mind paying this part because you’d have to do this out of pocket anyway and letting Amazon do it has its perks, including Amazon guaranteeing a certain sell volume.

The downside of this though is that sometimes Amazon demands a certain volume from you for your product, and you have to do it even if you don’t think it makes sense because they think the demand would go in a certain direction. When it doesn’t happen, Amazon asks you to get it off their hands and figure out how you want to offload it. You eat the FBA costs and the costs to get it back, on top of any other costs you may incur while getting rid of this inventory.

This system works for some sellers, but not for others who can eat up these losses.

This only describes one model of selling on Amazon i.e. as a vendor. The vast majority of sellers are not vendors and manage their own inventory quantities. The way the system qorks changed several times over the years, and was notably very constrained during COVID, but these days Amazon just provides a recommended qty per sku and an overall hard limit on total square footage each seller can use. The limit is determined by sales volume, so higher volume sellers get more space. If product doesn't move, the seller pays storage fees and eventually (more expensive) long term storage fees.
Former Amazonian here, great company and all that, I enjoyed it.

I felt I can provide data here, FBA is not the end of the road, there are more levels so to say, FBA and prime etc.

Now, FBA has some potential pitfalls.

- VAT, Amazon will often store in a country where its most economical for them, ie, for German sellers, there will be a warehouse just after the Polish border etc. They will do this unless you explicitly disable this. This leads to very complicated VAT fillings and correction

- Lost, stolen, damaged cargo. If your merchandise gets damaged or stolen etc, its not so easy to get new storage space allocated/assigned.

- Recalls, if your ASIN is subject to a recall, god help you, if you use amazon, you will have penalty points on the account health page, affecting many other things. All and and FBA ASINs affected , whether youre selling it or someone else need to be removed. The recall is only finished after that. And youre paying for the storage or recovery or disposal.

- Shipping confirmation. You are supposed to inform about any package and order status very soon after the order is placed. Same for tracking status. Now, sometimes the Amazon api for that is buggy and you get heavy markdowns on account health even though you provide the data, just the feed dont work.

Still, when its all said and done, Amazon offers the best product and logistics if you just want to sell your product.

There is no way any competitor can offer the same quality at a better price.

Keep in mind, whatever the percentage is, the ecommerce business has a margin in the single digits, all of Amazon is subsidised by the money making machine AWS.

Why did you leave? low margins?
Tax on the stock vesting(vesting is generous but taxed at 50%) and a disupte about the location for remote work made me look somewhere else.

I have to say the working hours are very nice if you are l4-l6, so is the compensation and all you have to do is clock in and out on the second(no joke, even too early can lead to penalties) and if you are proactive with feature improvements and not shy to do presentations, your promotion will be activelly pushed. Even if you want to change departments entirelly, there will be no hard feelings, you can announce that without hesitation. You can go to l4 from l3, as l4 is where the vesting starts. And no need to be the superstar as far performance goes, just dont be last. If you are last, youre still safe from a sack and a pip does not really happen, theres always someone to help, aka nesting team or something. If you dont perform, its on them. Pick people for help as early as possible and you will be golden.

Of course there are some rather annoying people there too, some even seem to have an allergy to work, but all in all, its bright folks and a good environment.

Once you habe been there as l3, l4 or higher, due to the rather steep initial learning curve(all paid and well designed, even later on, you can chose upskilling, also all paid), you are always welcome back, believe me that, I best not go into details.

I might go back some day, I think it is the best option if you feel like working in a big corporation for a while.

I have accepted an offer from a startup paying much better for the time being, but I always keep Amazing in the back of my mind.

Keep in mind this only applies for l3 to l6, I cant speak for the rest, but once youve been there, you will see why they always welcome people back, the onboarding cost and training for the first few months are vast. I always felt sorry if someone out of amazon would be hired for something like fba dept manager. But yes, seen many people have a one year break without a promise of going back which just felt like they wanna be a stewardess on planes for a while.

Also, this was in Europe, where many employment laws have to be followed to the t.

I would give them a try anyehere in the world, learned a lot there, including how to create procedures for things which have none, theres some pretty important things you can start from scratch, be the creator. They call it taking ownership, which is always rewarded, even if it takes you a month. You get these accolodates and stars and things.

I think the complexity is around the 'Prime' flag, which has a huge impact on sales and can force you down the FBA route if you actually want to sell in volume.

('Seller Fulfilled Prime' is possible - however currently closed for new applicants, is more targeted towards very large shippers, and there is pressure from account management on utilising Amazon Shipping as part of this at least in the UK)

Yes, you need to "earn the dues" before they enable that, I think its for good reasons, they protect the product and I would not advise a newcomer to even try prime FBS if they habe a certain volume and a very professional infrastructure is needed.

They have very protective rules for sellers who sell jewlery and such, too.

This is also a pretty normal rate for retail. If you want to sell your alpaca mittens at the tchotchke shop on Main Street they’re also going to want a wholesale price that’s 50% of retail.
100% retailer markup is the norm, no?

and also, they made their name as a discount bookseller and still have the best price on books, correct? So presuming they're not marking books up 100%

More like 200%-900% on a lot of items. When I was a sales associate at big box retail the big ticket items were sold nearly at cost (~10%). The addons were where the money was made and fat commissions. If you were selling Playstations the charging stands, USB cables, HDMI cables cost the store a few cents but marked up $10-$50.
I use to work at Radio Shack in college. You would be surprised by the retail vs wholesale prices of accessories and small ticket items.

But simple question. How much do you think movie theaters and restaurants pay for food and especially fountain drinks?

I'm just surprised some MBA hasn't reinvented company scrip yet. Maybe they'll try it internationally first, Walmart was using scrip in Mexico until 2008.
If it's not worth it, why are sellers using Amazon and not, say, Shopify?
Because many of the most successful sellers either own dozens of products, and have spent years optimizing their listings and winning the trust of the algorithm, or are direct fronts for Chinese manufacturers, who don't have the cost of the middleman. In either position, selling at volume on Amazon is profitable, and the eyeballs are there.
So it is worth it, then
In an interesting sense, this is because of Apple.

It used to be simple. You set up an online store; you put some ads on Instagram and Google; you earn back what you paid for the ads quickly. With each additional sale, targeting of customers gets easier, and cheaper.

Apple dropped their privacy changes. Don't get the idea that they had good intentions and are protecting the user- they've got an ad product on the way and just wanted to raise prices across the market.

So now in order to get those first sales, you need to put your ad in front of a lot more people because you can't tell which ones will be interested. That raises the cost of acquisition for your customer base, often beyond what's economically viable.

So who wins? Amazon, because they own a massive customer base and if you pay them enough, they'll put your product in front of customers. They have market power and they happily take advantage of it.