The difference is between retailer and marketplace. Both Amazon and Costco are retailers, and both could use that knowledge to decide what products to self-source for better retail margins. Either way, still a retailer, but maybe also a manufacturer / wholesaler.
But Amazon is also a marketplace. In that role it acts as a "rentable retail space". Using the data of the retailers in your marketplace to decide what to make/wholesale and then retail is another layer.
You could easily argue that it reduces to the same thing. But societally we've excepted that the retailer is a full layer in the system and gets full access to the data flowing through it. The marketplace itself is historically more of a fee-for-use type of thing, so its not an ingrained concept for us.
It's hard to make new laws to address specific problems with Amazon and at times it acts like both; sometimes it's mostly a marketplace and other times it does things like mixing inventory that make it more like a retailer (in the sense that they're presenting a unified front for multiple suppliers).
You're misreading the ruling. The ruling is for state law not federal and says the following, "The Appellate Court didn’t agree with Amazon’s stance. It noted that the product had been listed on Amazon, was stored in an Amazon warehouse, had payment facilitated by Amazon, and shipped it out in Amazon packaging, proving it to have a hand in getting it to Bolger and thus liable under California law." So, this stance will change depending on the product and what law is being alleged to have been broken.
This explanation makes the most sense to me, but there is something that you are leaving out. Not only is Amazon a marketplace, but it is THE marketplace. Amazon has an effective monopoly on small-seller logistics and marketplace services in the United States (and many other places), soup-to-nuts.
If you are anything other than a massive corporation, any manufacturer that chooses not to sell through Amazon and utilize all or most of its services (marketplace listing, payments, warehousing, delivery) will be at a massive cost disadvantage and will not be able to compete with other sellers that do choose to participate with Amazon.
And more significantly, perhaps, if you don't sell through Amazon's marketplace, you are often unable to compete with Amazon itself.
This just isn't really true. It depends on what type of product you're selling, but there are a huge number of independent ecommerce stores that do extremely well.
I sell dog treat mix (coopersdogtreats.com) - I do much better both in terms of margins and overall sales on my own website (with traffic coming primarily via paid FB ads) than on Amazon.
That's not even including other huge marketplaces like walmart.com, Chewy, Etsy, etc.
Amazon doesn't have a monopoly on small-seller logistics - I'm about to move all of my logistics over to a 3PL, and there are plenty that will cost-effectively work with startups (ShipBob, Shipmonk, etc. - just Google "ecommerce 3PL" and you'll see what I mean).
How much Amazon plays into your business obviously depends on the category, but the idea that it's impossible to compete in ecommerce unless you're on Amazon is an easily disproven myth.
What people kind of miss is that outside of FBA it's the actual logistics companies that are handling the logistics for all merchant fulfilled Amazon sales. The normal logistics companies are also handling a ton of the work involved with getting goods to FBA. There are lots of 3PL solutions that can deliver FBA style performance at a lower cost to the seller. There are other advantages to not using Amazon as a seller or just using Amazon as one additional channel among multiple ones.
I've seen at least one other comment saying this, and not that I don't believe you, but can you name several companies that do that and come within say 150% of Amazon? Last time I really dug into this was years ago and nothing was really close then, maybe it's better now?
My other question to comments :
> What are the Amazon level logistics and delivery service that is available to a smaller retailer? I see some nascent choices but they really aren’t that close.
I've priced out several and all are within 150%. Shipbob is the closest at 10-20% more expensive than FBA, but depending on your product it may actually be more cost effective. I need them to pick and pack several items, since I sell my stuff individually as well as in kits (boxes of multiple items, which are vastly more popular than people buying individual items), and they include most of the cost of building out those kits in their shipping fees.
I'm happy to read what you are saying, and stand corrected.
I wonder, though. Do you think your experience is typical?
Am I wrong to think that your product is more niche and premium than the most products that are sold via the Amazon marketplace? Do you think that you might get more repeat business than most products sold on Amazon?
It's been a while since I've looked into pricing for FB ads, but my sense is that a product with more narrow margins and less potential for repeat business could find it difficult to attract customers via advertising without increasing prices beyond what could be found on Amazon for competing products.
I think Amazon's too big and broad to describe any experience as typical, but there are certainly differences in the type of business.
You are 100% right that my products are niche and premium, and that definitely makes a difference. But on the other hand, if you're a startup selling a commodity product, you're in a bad position for a whole lot of reasons other than Amazon.
From an advertising standpoint, you're right - I have the advantage of selling a product that has a high repeat rate, and that's helpful. On the other hand, my AOV is fairly low and my margins are okay but not extraordinary.
I think that the points you're raising here are what matter - not Amazon. If you're in a business that is one of: high AOV, high margin, subscription/frequent repeat customers then you're okay. If you're in a low margin, commodity business and you aren't operating at a huge scale, you're not in a great spot.
Ultimately, I think the actual value added by Amazon for startups is trust - I know that if I order from them, your product will arrive on time, and if there's an issue it'll get fixed ASAP. All of the other stuff, like two-day shipping and customer-friendly return policies, is doable off of Amazon. Even trust is achievable in other ways, though - my company was featured in an Associated Press article that was broadly syndicated, and when I slapped "AS SEEN ON USA TODAY, FOX, KTLA AND MORE" on the top of every product page and the top of my FB ads, it made a huge difference immediately.
And actually now that I've typed that I'll add one more thing in Amazon's favor - ease of use. If you don't know a lot about ecommerce, it's pretty easy to get set up and ship them product. It's also easy to advertise, simply because their advertising is much, much simpler than Facebook. There's no copy and limited ability to use creative outside of your product images, so you don't have to constantly test stuff. While my sales on Amazon aren't high margin, they're also extremely low effort.
I'm curious about how you would compare Amazon vs your own site. Things like Amazon's buyer-friendly return policy comes to mind...which clearly come at a cost for sellers (and buyers, fairly directly).
I have a very small number of returns on Amazon, since my product can't be used and then returned.
Amazon isn't terrible, but it has two big problems for me. First, margins are lower because I'm not only paying them a cut of every sale, I'm also paying for Amazon advertising. Second, I don't have a relationship with the buyer, which is the real killer. I do really well with repeat sales via email, and those are where the real margin is since I'm not paying to acquire those users again. On my site, I can afford to break even (or even lose a little money) on the first sale, where I must make money on each sale on Amazon.
I will say the one big difference between my product and many on Amazon is that it's not something that you seek out - very few people are searching for "dog treat mix" on Amazon (or Google/Bing/etc.). Amazon does have some types of advertising that work well for targeting other luxury dog goods, but my volume is going to be limited there. Facebook advertising works better for me, because I can target people by demographics and explain the product in an ad.
To that point, it may be important to be on Amazon if you're in a commodity business. On the other hand, it sucks to be in a commodity business for so many other reasons that I hardly think it's reasonable to pin the blame on Amazon for the difficulties there.
The one thing I will say about Amazon vs. Shopify that surprised me is that Amazon's support tends to be quite competent (at least if you call them - you get nothing but canned responses via email), while Shopify's is just terrible. They can help with basic issues using the software, but when it comes to real problems, like bugs in reporting (or more recently in my case, a bug where they undercharged a customer), they just say they'll get back to you and never do unless you are incredibly aggressive about hounding them.
I'm still not understanding the distinction here. Costco and Amazon both sell company-brand products, alongside non-company products. Costco and Amazon collect and analyze sales data from the sale of both company and non-company products.
You have the wrong mental model. Amazon isn’t Costco. Amazon is a shopping mall that has access to its tenants’ sales information and also owns an anchor store in the same mall.
Costco can determine that Best Brand shoes sell in its stores and decide to source shoes themselves and stop carrying Best Brand.
Amazon can determine that the Footlocker in their mall is making a killing selling Best Brand shoes and either sell Best Brand shoes in their anchor store or source their own shoes, all at a price that Footlocker can’t match. They can also advertise those shoes throughout their mall and change the layout so customers have to walk past their cheaper shoes to get to the Footlocker.
This explanation doesn't explain the difference to me at all. Costco can see that vodka sells well, so they can make Kirkland brand vodka. They control the layout of the store, so they can change the layout so people see Kirkland brand first when they get to the liquor section. They know the pricing of other vodkas and which ones sell well, so they can source Kirkland vodka at an ideal price point.
None of what you've described about Amazon differentiates it from Costco at all.
It's simple, Costco actually buys the inventory, and resells it. There are more advanced kinds of agreements (like buy back X number of coats of you sell less than Y) but that's largely how they operate.
Amazon is much more like a digital mall in that they rent out their store features for sellers.
However, Amazon would like to be seen by the customer as Costco and this causes dissonance between how they treat sellers, and how they treat customers.
Okay, you've described the business models, but so what?
In both cases, they still use the sales data available to them to create and sell their own products at the expense of third party sellers. In Amazon's case, that means diverting customers away from third party listings to their own. In Costco's case, that means making fewer purchases from third party sellers because the amount of shelf space available decreases as Costco puts more of its own products out.
The type of sales relationship that they have with third party sellers doesn't change the fact that both are using sales data to create and sell their own products at the expense of those third parties.
S/he may argue that public officials manage to do it too, despite its illegality, hence it's Okay for Amazon to do so. For some, whatever Amazon does is okay, there will always be some other cases happening elsewhere that in some twisted ways are the same. I think most people are only ever consumers, and can't escape their subjectivity on this matter. They get theit toys delivered within a couple of days, the invoice says free delivery, they want things to remain that way so what ever Amazon is doing is okay (for them, for now)
You can't see the difference because you're conflating the terms seller and vendor. There are no third-party sellers at Costco.
The difference really comes down to three things: the layer on which the competition occurs, whose data is being used, and whether the competition is fair (which comes down to risk/reward).
At Costco, the competition occurs between manufacturers. Costco uses its own retail data as a retailer. The risk to the vendor to sell at Costco is marginal (even if they have to pay for shelf space) compared to being in the market at all. Costco can compete with a store brand, but if they want to sell the vendors products they have to go to the vendor. They can't undercut the vendor with an equivalent product unless they make an equivalent product for cheaper. The risk is spread out and Costco owns a healthy amount of it.
At Amazon, the competition occurs between retailers. Amazon uses its competitors data, which it gains by being a 'marketplace'. The other retailer carries all the risks associated with being a retailer. Amazon takes a piece of it, the size of which depends on what 'services' the retailer uses through Amazon's marketplace. If the retailer fails, Amazon loses nothing. If the retailer wins, Amazon can use the retailers' data and begin selling the same product. It can use its size and the retailers' own numbers to get a better deal with the vendor and undercut the other retailers price on the same exact product. (Store brands are also an option and an issue, but if you're focused on that, you're missing the forest for the trees.) The retailer takes the vast majority of the risk, but Amazon can, at its option, swoop in and take the majority of the reward.
In the Costco model, Costco pays up front for the vodka and incurs all the risk if the vodka doesn't sell. In the Amazon model, Amazon charges third-party sellers "rent" to be on the platform in the form of a cut of all transactions, but the third-party sellers still incur all risk for inventory that doesn't sell. And then Amazon turns around and uses the sales data from the third-party sellers to undercut them later on.
I am not sure about Costco, but Best Buy does not purchase all the items it sells. For some items it only remits a payment to the manufacturer after the item has sold. Should they not get to see the sales data?
Not always true. Usually, Costco/Target/Walmart/etc. pay for it before it sells, but it’s not always like that. “Vendor” products like Frito Lay, Bimbo, etc. are sometimes given the ability to just put product on the floor at no cost to the retailer. When it sells, the retailer will send money to the vendor.
Consignment sales represent less than 1% of retail sales at these stores, and generally only are required for new products that the retailer will not purchase in bulk before the distributor proves market demand.
Generally, for the sales you have described, the consignment sales are paired with marketing efforts by the distributor to demonstrate customer interest. If the test succeeds, the store will purchase future lots from the distributor. If the test does not, the product disappears from the shelves and the distributor stops selling it.
No, those are Costco’s suppliers which are completely invisible to the customer and completely different.
Costco purchases the items from the suppliers, generally. They then resell them to you for a markup.
With few exceptions, every single item in Costco is sold by the first party seller: Costco. So my point stands: third party sellers are generally not a part of the Costco experience.
Yes, but the terms of the relationship make Amazon's behavior anti-competitive. Costco isn't going to various brands and asking them to sell at Costco, they buy inventory and resell it at a profit. Amazon has convinced a large portion of the retail market to feudalize themselves on Amazon's platform, then they are using the data accrued to take over the markets of their current and former tenants.
Costco sources Kirkland products from people they already do business with. So if they see your Vodka is doing great, they call you and ask (well, as a giant customer of yours "ask") you to make a Kirkland version as well. On those bottles you'll make less, but you're still going to make a lot of money there. And getting cheaper people to buy your product without diluting your brand. And a lot of money off your branded product. Most things I've seen from manufacturers online is that being asked to supply Costco with their product and the Kirkland version is a win from both asks.
Unlikely? I literally just described an example of Costco doing this right now with vodka. That's not a hypothetical - Kirkland brand vodka is a real thing.
Your logic is backwards - when space is at a premium, making the most profit off each item in that space is critical. It makes more sense for Costco to do this than it does for Amazon, not less.
Store brand products are not new and not the issue. Let's pretend that Costco didn't currently sell vodka at all. If Costco wants to know if they should start selling vodka or bottling their own vodka and selling that they don't have access to the sales data from the liquor store next door to Costco. Amazon is letting other businesses take the risks and using sales data from those businesses to outcompete them.
That's the right way to look at it, missing just one crucial detail.
Costco pays for the items that appear on its shelves (excepting the <1% of goods that are on a consignment basis, usually new trial products). The distributor of the store brand and the name brand have already been paid for the products (and are usually the same company).
Amazon gets paid by the distributors (aka third party sellers) on Amazon.com for handling their sales (even in instances where it is not handling fulfillment), while also competing against them. That is where the anticompetitive concerns arise.
If Amazon just sold stuff through the Amazon.com seller, and didn't have third-party sellers, (or if it operated a separate website for third party sellers) that would be fine.
Don't most large brick and mortar retailers maintain refund for unsold goods agreements in addition to defects? Generally only exercised if they are complete failures.
Effectively the difference in practice is a matter of financing and grain of operation - older retail would gain more and give no extra to upfront sales of say toliet paper after a demand spike raised prices while Amazon would give them a per sale percentage cut.
At what point does own involvement in consignment sales models become not fine? If it works at 1% consignment. Is it 25%? 50%? 75%? 90%? Or more likely it doesn't exist because the whole concept is a fabrication that pays no attention to real law and operates in the court of public opinion to push their bullshit which wouldn't even need a defendant motion before winding up dismissed by a judge because they cannot point to any real laws?
I don't know why HN is so focused on consignment sales. They are a tiny portion of the retail market, because no store (even Walmart) has the leverage to force consignment terms on their suppliers except for the tiny portion consisting of new products that distributors are trying to get stores to start selling.
Also, Amazon does not give a per sale cut of the percentage, nor would they, since that's not how consignment works. Consignment goods are still sold at retail price; the only difference is that the supplier only gets paid for goods that sold. They don't get to share in the (additional) profits if the retailer (aka "Amazon.com")charges more due to spiking prices.
But Costco does sell Vodka and wouldn't have made that product if Vodka sold 10% of what it actually sells. Because Amazon sells literally everything, is it a crime to do what Costco does, just on a bigger scale?
The only end-goal that would actually solve the problem fairly is if companies couldn't sell first-party products (or products from a partner where they have a vested interest in) in their store. If you just take care of the one company, you end up with other companies doing the same thing in 20 years like how iOS still has a default music player when MS got burnt for that with having a default browser.
I feel like people in this thread are being deliberately obtuse. The issue isn't about selling vodka. It isn't about selling store brand alternatives to brand name products. The issue is where Amazon is getting their data from when decided what products to sell and what store brand products to make.
When Costco decides to make a store brand alternative they are using sales data for things they have sold in their store. Amazon is using data for things other people have sold. Amazon is not doing what Costco does.
This is not about making a home brand. Amazon can literally look at the sells data of e.g. some seller which sells Nike Air Jordans (as a stupid example) and go and source those themselves and offer them (the exact same brand) cheaper than the seller, because they have all the sales data. Now how would Cosco do this?
When Costco sells Kirkland brand vodka and Grey Goose, all bottles of Grey Goose are sold the same way. Costco purchases directly or from a distributor. Grey Goose may or may not pay for shelf space, IDK. Purchasing premium shelf space is a common practice for other retailers and grocers. When Grey Goose is sold on Amazon, either Amazon buys it and re-sells it or Grey Goose sells it through a marketplace account incurring Amazon's marketplace fees. So Amazon can always have two listings for one product, a marketplace listing and an Amazon listing. Their Amazon listing can always be cheaper.
Maybe Costco is different, but at a lot of stores the line is blurrier. You might be able to get your product on store shelves, but if you want good placement, you basically have to rent that shelf space.
> Costco and Amazon collect and analyze sales data from the sale of both company and non-company products.
It's similar, although personally I think the relationship between the companies is meaningfully different:
Costco purchases product from manufacturers, and may choose to source product from other manufacturers (including under its own brand name). It uses it's own sales data to make this decision.
Amazon acts as a marketplace for other businesses to list and sell their own products. These businesses are online retailers which use the Amazon platform, and pay Amazon fees for this service. Amazon is then using other retailers sales data in order to inform it's own business.
The difference is with Costco it is their own sales data, while in Amazon it is the sales data of other retailers. It would be an issue if Walmart had access to Costco's sales data and not visa-versa (this would provide Walmart with an unfair competitive advantage). Similarly other smaller online retailers do not get access to Amazon's sales data, but Amazon get's access to the other retailers sales data who use their platform, and will then use this to compete with them.
Despite this I don't see how the case is that Amazon is being anti competitive while Costco isn't - just because they purchase and resell inventory doesn't mean Costco (or Walmart or Sams Club etc) doesn't hold the same power over their product suppliers that Amazon does to do data science on their sales to determine what new products to make in-house.
Plus, walmart is now a marketplace as well. This overpriced GPU is 'Sold & shipped by Monoprice Inc'. It's only a matter of time before Walmart commits the same anticompetitive acts as Amazon using Marketplace data. https://www.walmart.com/ip/Zotac-NVIDIA-GeForce-RTX-3080-Gra...
Analyzing their own sales and analyzing other people's sales on their platform are two different things. Saying that Walmart may also try to do this in the future does not make it right or excusable.
Who takes the risk in each scenario is different. If Amazon was only using their own sales data then they'd need to start selling a more diverse range of products themselves to acquire that data. Some of those products would sell well and some would be flops. Amazon would have invested money into selling all of them so the cost for the flops negatively affects their bottom line.
In the current scenario someone else sells things on Amazon and is taking the risk that the item they're selling will not sell well. If the item is a hit, Amazon swoops in and starts selling it themselves or possibly makes a competitor and sells it themselves. Either way, Amazon reduces their own risk of selling poorly performing products while also cutting into the upside for the vendor who took that risk when they are successful.
Edit: I forgot to mention above that the people taking this risk are paying Amazon to do so.
> Just because they purchase and resell inventory doesn't mean Costco (or Walmart or Sams Club etc) doesn't hold the same power over their product suppliers that Amazon does
With the Costco example, Costco as a retailer holds power over their product suppliers. If they are making an own brand, what they are doing is buying it from another supplier and asking that manufacturer to put their own label on it. These relationships can be anti-competitive and present opportunities for market-abuse, but in a meaningfully different way to the Amazon example.
With the Amazon example, Amazon holds power over other retailers using their platform.
And with the Walmart example you have stated, I do think that suffers from the same problems.
To be competitive, Amazon would be able to say everyone has access to the same information to rent out the online marketplace's booths. This is the issue at hand--Amazon has all of the information and is breaking its own rules in using it to give them advantage in the marketplace. Greedy greedy capitalistic cheesey.
When Costco sells a product they’ve already bought it (a retailer) on the other hand when Amazon sells something they’re just acting as a middle party for the item in most cases.
That's often not the case. Retailers like Costco, Walmart, Best Buy etc have a wide variety of different selling arrangements. I've sold products through all three and often did so on terms that gave them full right of return for any unsold product as well as significantly delayed payment.
This combination basically nets out to be financially the same as pure consignment. They won't pay me for my product until well after it has sold-thru to an end user. Everything that's unsold comes back to me (and they bill me for shipping both ways!) In the meantime, all I have is basically an "IOU" promise to someday pay IF it eventually sells (and they always drag out the payment beyond the already-extended due date).
Also, if I want to be featured in their circular I have to "buy" that just like an ad in a magazine except the retailer will (usually) DFI (deduct from invoice) the "ad" cost, which means they just owe me less (if and when the product sells and they actually pay). The same is true for getting my product displayed on an end cap or with in-store signage.
The big retailers bring in new products to "test" all the time and do so at basically no financial risk to themselves (other than the opportunity cost of the shelf space) while capturing all the sales data.
I'm not sure how Costco works specifically, but that's not the case in all retail. At GameStop, game publishers only got paid when their games sold not when they were put on the shelves. If there are 100 disk cases put on the shelves for Call of Battlefield and none of them sold, eventually GameStop would return them and the publisher got nothing.
All of Amazon's "we're not actually a monopoly" and "we're not responsible for defective products" arguments are based on this. They claim they are very much NOT a Costco or Walmart.
Amazon sells products and facilitates others' sales.
If Amazon carried the entirety of their inventory themselves like Costco or like Walmart used to be (before the expansion of their own online marketplace), it would be a distinctly different situation.
Costco buys products and resells them. Costco's research numbers are paid for entirely by Costco.
Amazon rents space to merchants where those items are sold through the site for a fee. Amazon is never on the hook for a sale and is basically getting paid to do the market research to set up as a competitor.
Yes, Costco does do some referral sales but I can't think of anything which has gone on to be a Kirkland product.
I think the difference they are getting at is something akin to this.
In one situation you run a stall and buy products from people to sell at that stall. At some point you use what you've learned doing this to sell your own product.
In the other, you don't buy anything from anybody. Instead, you rent out a stall for other people to sell things from. You then watch the stall and use that information to open your own stall.
The first case seems pretty normal to most people, I think. The person you were buying from originally doesn't inherently get some kind of assurance that you will always buy from them in the future. There's no difference to the seller if you buy from somebody else, don't sell any of that product, or make your own. We just don't expect that buying goods from somebody inherently adds any other kind of obligation. It's two equal parties making an exchange, and nothing more.
The second case, however, I think is not so clear cut. All of the sudden you have a lasting relationship between two unequal parties. These are the sorts of situations where you tend to find more implicit or inherent obligations on the participants. It's no longer the guy you sold that thing to not buying from you again, it's your landlord competing with you.
I'm not trying to pick a side here, so much as I am trying to explain why people might not see the two situations as identical. And of course there are plenty of real-world complications too.
So how do you factor in the recent ruling in California?
> An appeals court in California has ruled that Amazon can be held liable for products sold through its marketplace by a third-party seller
That seems to obviate the distinction of 'marketplace facilitator' vs. 'retailer' for them. That makes them much more like Costco with a special relationship with vendors to set up vendor-specific sections in their store.
Personally, I've always detested the 3rd-party market in Amazon and wouldn't mind seeing it go away.
Amazon doesn't see it that way, their optimised their business as an ad platform + logistic solution. Higher margins there than bothering acting as sellers with lower and not even guaranteed margins.
Amazon doesn't see it that way, their optimised their business as an ad platform + logistic solution. Higher margins there than bothering acting as sellers with risky margins.
But Amazon is also a marketplace. In that role it acts as a "rentable retail space". Using the data of the retailers in your marketplace to decide what to make/wholesale and then retail is another layer.
You could easily argue that it reduces to the same thing. But societally we've excepted that the retailer is a full layer in the system and gets full access to the data flowing through it. The marketplace itself is historically more of a fee-for-use type of thing, so its not an ingrained concept for us.