I feel that this is an untenable situation and business model. When people first took to buying their "random crap" via amazon, it was at least backed by an amazon customer service experience that defaulted to very good - amazon prime shipping, consistent shipping and packaging, consistent return policies.
But once these marketplace sellers came into being, the qualify of the experience is on the whole, lacking.
Years ago, my family would use a particular salad dressing enough that when we found it on Amazon as a "send me a case of this every three months" thing, we were happy with that. It was marginally cheaper than buying at the store, but it was far more convenient
At some point, we started receiving things from sellers that were obviously not Amazon themselves, and then endless emails asking us to rate their service, etc. Sometimes the product would be poorly package or damaged in shipping. Once it seemed like something had maybe turfed off their pantry shelf into a box to send us. It was annoying but not worth our time to 'fix', so we just canceled that order.
But lately it seems like things have been getting harder to deal with when these third party sellers are involved. I don't want to save $1.37 on a $200 router that I receive in obviously opened-and-returned condition and missing parts, and then receive an insulting letter from the (actual) seller about it somehow being my fault and saying that I should pay the return cost.
It may be possible go get amazon involved in resolving this - BUT IT IS NOT WORTH MY TIME. At some point, it was easier and cheaper to use Amazon for all these types of purchases, but Amazon Marketplace is hurting their brand. Weakest link and all that.
(Maybe we'll get a return to brick & mortar stores for radio-shack or similar things.)
I'd love to see an "Amazon Only" mode that was a switch on the main page to only show the store for sold by or fulfilled by Amazon. I'm sure this can be done somehow and would make the experience much better. The crap 1000 review emails would be nice to disappear as well.
You can choose to filter by items only sold by Amazon. You can also choose to filter by "Prime" items only- with a few small exceptions, these are only items sold by Amazon, or Fulfilled by Amazon (ie: the item is physically in Amazon's hands, but sent to Amazon by a seller).
Bias note: I've worked in Amazon Operations Tech for the last 4 years.
Filtering by seller requires you to have selected a category, which I find you often have to do several times over to cover all the categories in which the item you're looking for might be located.
Filtering for Prime also includes anything fulfilled by Amazon, which I think most people will agree only marginally reduces the risk of getting sub-par quality from random seller number 5.
I am yet to find a reliable or convenient method of limiting search results to Amazon-only products.
The lack of a switch like this is especially annoying for people who live in "other countries". Like Sweden, in my case.
Amazon gladly ships here (from UK, France, Germany, Spain, US, etc) - with quite decent product and shipment prices.
However, all the search results are polluted by these non-Amazon sellers' results, who 90% of the time don't even have an international shipment process.
In the end, this just makes me avoid Amazon and go somewhere else.
This is a bit off-topic, but here is another stupid thing. My experience with amazon.es:
1. I go there. Everything is in Spanish, despite my IP being IP-geolocated to Sweden by every single IP-geolocation database I know of. My browser's Accept-Language HTTP header is "en-US,en;q=0.8,sv".
2. I manage to sign in with my global Amazon account - they can now see that my shipment and verified payment location is in Sweden. Everything is still in Spanish.
It would be fine if it was just the product descriptions which lacked english translations that were not available in e.g. English, but this scheme just seems incompetently managed. They run each national site as a silo, language-wise?
This kind of thing was a fun adventure ~17 years ago, ordering exotic stuff from Amazon Japan, but in this day and age it just feels stupid.
Just to provide a counter-point, I'm the exact opposite. I get annoyed when websites try to change the language for me based on my location. Just because I may currently be in Spain doesn't mean that Spanish is my preferred language. When I go to amazon.de, I expect everything to be in German. If I need a different language, I prefer to choose it myself. Even having the language connected to my account wouldn't be preferable to me. I would still prefer the site to be in the language of the "hosting" country (but I'm sure this is a much less common preference).
I guess my point is that different people have different preferences which may be one reason why the site isn't designed like you would like it to be.
I understand what you're saying about how you expect amazon to be one unified shopping location. But I imagine that amazon.es is designed primarily for the Spanish market and people in Sweden using the site is a much smaller, unintended market.
First: I specified that I was providing them no less than three separate data points: location, language, shipping address.
I wouldn't expect some tiny spanish company to do this.
However, with a global giant like Amazon I do. Especially when they have already solved the hard part - the logistics around international shipment and payments.
The sensible thing for amazon.es to do when they detect a customer who is probably not from Spain (based in e.g. IP-based geolocation) would be top show some dismissable banner at the top of the page saying "We see you are visiting from Sweden. Would you like to read the core part of the site in English? <YES> <NO>")
Yes. Click on one of the categories on the top of the left sidebar. Scroll down, Amazon.de is listed under "Anbieter" (provider). Click on Amazon.de, and you get your search filtered. You're welcome.
Keep in mind that Sold by Amazon and Fulfilled by Amazon are two very different things. The latter just means Amazon is providing the distribution service; you're still buying from a third-party Marketplace seller. (I believe sellers are required to follow Amazon's return policies when it's Fulfilled by Amazon, so there is that at least.)
Regarding a display switch, you can do that in the search results listings as mabbo described. It would be nice if there were a global setting like you're talking about that applied to all pages and not just search results.
Any seller is required to follow Amazon's return policies, not just FBA ones. And Amazon will support the buyer in most cases. If the seller refuses to honor a return, file an A to Z claim.
I agree fully. Add in the blatantly fake reviews that are plaguing Amazon (my wife recently got screwed by this on a purchase, thinking it was highly-rated) and Amazon is losing what, IMHO, is a core piece of what got them to where they are. Namely, trust and reduction of risk. They had great reviews that were actually helpful, and if there was an issue they'd take care of it. Increasingly, this is not the case, but it may matter less for them now that they dominate the market.
Today, I increasingly make major home purchases from Bed Bath & Beyond of all places because of their policies. Even if I order it online, I can have it shipped to the store and return it in store for free, with a damn near unlimited return due date. I've even returned a pillow I slept on previously, no questions asked. Add to that the fact I can physically look at the item (absolutely critical for determining color match, build quality, etc.) and for some purchases Amazon just doesn't provide the necessary experience yet (despite having a retail location now).
On the other side of this story, people trying to be Amazon resellers are plaguing every entrepreneur-related forum I used to visit. Between that and subscription boxes they are clearly the next big "Get Rich Quick!" scheme. I can't wait till they die off.
A video can't show me how a fabric will look under certain light, or how it will compare against my rug.
A video can't tell me that while the piece of furniture looks nice, the drawers are really cheaply made and don't slide nicely.
A video can't tell me that what looks like a nice finish actually feels like cheap plastic.
If you have suggestions for how to tackle that, I'd be very interested. There's some things though I just can't see myself ever buying online, and unfortunately retailers are trying to shift more and more of their purchases there without the benefits of things like free in-store shipping/returns. If they did that and I could see it in-store, and return it free, I'd totally do that in a heart beat. I bet the shipping rates and warehousing costs would be low enough to be worth it.
Video / images can be reasonably good, but only in a subset of cases. I find that there's frequently no real substitute for seeing the real thing. There are reasons Amazon initially focused on books: highly uniform, online competes favourably with storefront on selection, quality is not usually a distinguishing element.
Except that sometimes it is. I'd recently ordered a book by an author I've just discovered and like greatly. When it arrived (ordered through a bookstore), I was disappointed to find it was printed on pulp rather than acid-free paper. Small details can matter, and if I'm investing in a physical artifact, I'd like it to last.
That said, there's some low-hanging fruit / reasonable use-cases which might help direct your efforts.
I've often been hugely frustrated by how poor existing product descriptions and images are. Simply ensuring that accurate and useful information are presented is hugely useful.
Images need to be big, they need to be detailed, they need to be well lit, and if at all possible, they should offer 360 views -- rotation or front, side, and back shots, plus angles.
On that last: I've been looking at a particular piece of furniture, actually made several trips to storefront locations hoping it might be present. It's online-only. And the only view presented is a front-oblique, at about 30 degrees left and vertical. Yes, I get an impression of the object, but for a $1500 purchase, you'd think the vendor would invest just slightly more in presenting the piece. I've opted not to purchase it. Either having a piece custom-built, or refinishing an existing older piece, are preferable alternatives.
These aren't technical problems, they're process problems. Get vendors to address this shit.
The other place images are useful is for repeat purchases. If I've bought item X, then the picture's going to help confirm that I'm buying what I'd wanted. Here video's not so critical as just having clear product shots.
But as a general rule, I find the online shopping experience a strong negative.
Again to emphasise: most of the problems here are of behavior, not technology.
I'd strongly suggest you look up Akerloff's "The Market for Lemons". That details how information failures hurt all market participants.
And I'd again emphasise that there are areas in which image can help and more importantly, many more in which it CANNOT.
Pick your battles.
As for video: it's very high-bandwidth. What's the difference between, say, 15 seconds at 32 fps, and the coverage of, say 4-16 well-arrayed photographs. Square-on front, back, and sides, or a well-selected set of oblique shots.
Again: where catalog marketing (and online sales is just updated catalog marketing) breaks down is that the communications channel is very thin. I can tell far more in two seconds of handling a piece than I can in a minute of reading obfuscatory marketing copy.
>When people first took to buying their "random crap" via amazon, it was at least backed by an amazon customer service experience that defaulted to very good - amazon prime shipping
Unfortunately, there's a reason I buy what some people would consider "random crap" off Amazon -- I cannot find the actual item I want in stores. I can go to Home Depot, Lowe's, ACE, and several other hardware stores and spend the entire day, and NOT come home with the lightbulbs I'm looking for. Perhaps I can order them online and wait a few days for them, but usually I come home empty-handed, which is maddening.
Then there they are. Someone selling them on Amazon that even with shipping, is FAR cheaper than even the brick-and-mortar sites with their free shipping. Rinse, repeat for many other things, and next thing you know I'm ordering occasional "random crap" via Amazon since I can A) get it faster and B) cheaper through storefront sellers.
> "At some point, we started receiving things from sellers that were obviously not Amazon themselves, and then endless emails asking us to rate their service, etc."
I've noticed this more and more in tech, even outside of the retail ecommerce experience. I constantly get bombarded by requests to rate things and it's all getting to be way too much.
Seamless/GrubHub constantly wants me to rate every experience - I get that they want feedback and to ensure quality is high - but I'm not their QC department and I don't want to be.
At this point a pretty substantial portion of my email inbox is just "how was X? Rate X!" messages. Every damned startup thinks "oh but it's a quick thing" without realizing the sheer volume of it their users are being bombarded with from every which way.
Surely there's a better balance when it comes to quality assurance than "get the customer to do everything".
Paying actual quality control people - harder for some business models than others, but this has been the way it's been done for years.
A department store gets goods from potentially thousands of suppliers, and yet doesn't have to ask every customer to fill out a survey to do quality checks. Ditto your local transit agency can tell how well they're doing without polling every rider.
Obviously a harder problem when you never meet your suppliers, but IMO still a tractable problem - IMO part of the problem is that leaning on customers to do quality assurance has become a default rather than a fallback. Harder upfront quality checks also plays against the desire for hypergrowth (you can only vet so many vendors so quickly, plus some of the bad-but-not-horrible vendors are actually valuable for you).
Side note: it's probably more palatable for customers to give feedback if that feedback is actually acted on. One of the annoying things about, say, AirBnb ratings or Amazon ratings, is that they largely just feed into a giant pool of votes, and the only outcome of your feedback is an imperceptible nudge of some tiny fraction of a 5-star rating scale that's visible to other users.
So curated AB&B? That seems like you COULD do it, charge someone $50 to get curated by staff and then approved as a "4 star AB&B room". But what stops the place from dropping standards as soon as they get curated? Monthly surprise reviews? Seems like a really hard problem.
A retail store can check the quality of a pair of socks, approve it, then order a million of them. Maybe spot check a few, but assume most are the same quality. Can you ever really do that to AB&B rooms? Even hotels are spotty, the best info we can usually get on a given hotel is trip-advisor, which has all the problems of paid /shill reviews and annoying customers to leave feedback.
Retail faces many of the same problems - you can check the quality of a pair socks, but if you order a million of them there can be high quality variance and detecting that is non-trivial.
Also common is that you check and verify the quality of socks but 2 years down the line your vendor starts screwing you with lower-quality equivalents.
Quality control is a hard problem in any field.
> "But what stops the place from dropping standards as soon as they get curated? Monthly surprise reviews?"
That's what we do for restaurant hygiene - and at least around here it seems to keep a reasonably good lid on things. With the help of software it's also relatively easy to identify patterns of failure (as opposed to irregular lapses in quality).
> "Even hotels are spotty"
They can be, and this is a large part of why hotel chains even exist - they exist because quality information is really spotty and so the chain is essentially a guarantee of some consistent quality expectation.
Yeah, part of the problem is that many of these companies are middlemen and almost completely uninvolved in the actual product being sold (see: Uber doesn't manage any of its own cars, AirBnb doesn't manage any of its own rooms), and so far the pattern has been to pass the buck to customers with a "buyer beware" notice.
I contend this is not good enough - especially when many of these services are high-priced and advertised as substitutes for actually-curated (see: hotels) products with solid quality assurance. IMO in many of these cases companies need to do more for the margins they charge (e.g., Seamless charges 15% of all orders, but they don't provide any portion of the service, not even the delivery bits unlike Caviar or Postmates).
Ya for sure. To me this is why I am kind of leary of the Uber type business model. I get why it works, but it also seems so hard to maintain a level of control when you push everything down to driver responsibility.
The funny thing is Amazon was great at this for their original product line back in the days for books. They hired pro critics to write independent reviews. This is not emphasized anymore, probably due to cost.
I've started giving 1-star ratings to both the product and seller, for those that do this "email begging for a review" thing. Frankly, I regard it as spam, and treat them as such.
It's really ironic when a seller asks me to rate their item... when I already have given it a 5-star review + appropriate worded comment. It becomes a 1-star review, etc.
Saving less than 1% may not be worth your time in case of any problem. But often the difference will be much higher, and then it becomes a question of how much risk you're taking versus the savings.
One thing you can do is leave bad feedback, which encourages sellers to behave. Also, before ordering, just glance at the feedback for the seller. I doubt that your router horror story was with a seller with >95% feedback and >50 feedbacks. Takes a minute more and reduces the risk. Or, restrict purchases to FBA, which has customer service by Amazon.
I am increasingly using Google Shopping + Paypal. By using Paypal I can avoid entering credit card details and authenticate with 2 factor. The Google element gives user feedback and increasingly they are offering the "Trusted Stores" insurance. You also avoid the endless chaff of product lines that are unrelated to your current purchase.
I love eBay but have recently started using Amazon more and more, but usually because of prime (not a 3rd party seller).
You buy from 'some guy' on ebay, but I have complete confidence in buying on ebay (cause of the Paypal buyers guarantee) - in fact it's more risky selling on ebay than buying.
Yes, absolutely. If anything, Amazon is more strict on their third-party sellers than EBay is. But, there's a huge expectation gap between the service you expect from even a great EBay seller and Amazon itself.
I've built software for companies that sell on Amazon exactly the way these guys work. The products sold were different but the logic exactly the same. Write software that intelligently guesses the optimum prices and automates listing on Amazon. There's more to it than that, but this is the crux.
This is a TOUGH business to be in. They're under attack from every conceivable angle every single day:
1) Customers complaining on Amazon, often not legitimately
2) Minimum wage employees doing the "picking and packing" to ship to customers not turning up for work, making mistakes or worse still stealing from the company
3) The postal service letting them down not delivering an item which results in more 1) customers complaining
4) Suppliers spotting that my clients are on to a good thing and trying to rip them off, gouging them on prices or just messing them about in general
5) Margins so tight that they squeak when they walk. This business is all about volume of sales because of those margins. Which means you need to have lots of people, floor space, and other associated fixed overheads to achieve said volume
As if all of that isn't bad enough, you're in the pocket of Amazon. A company who at any given point in time for any reason they deem fit could simply shut your Amazon listings down and just like that your business dies.
I'm sure they're doing great, and I wish these guys well, but holy shit is this a tough business to scale.
Markets can only be competitive when information is ubiquitous and barriers to entry are low. This, I think, is one of the less talked about revolutions brought about by the internet: the creation of hyper-competitive markets that have led to superior outcomes for consumers.
You're not factoring in the fact that the Internet creates huge monopolies (just like money also has a natural monopolizing effect; in other words "rich get richer, poor get poorer").
We end up with a couple of huge Internet companies which will increase their pressure as soon as they are sure that they are "strong enough" (meaning: protected from healthy competition which usually favors the consumer).
Example: Amazon - they currently keep only a very tiny margin, because they are still about increasing their overall market power. Once they have maximized that, they will use their power to increase their margin and all players (including the consumer) will feel their heat.
That's not clear. The idea that (non royal-patent) monopolies always pursue dominance to corner price seems problematic.
Like Standard Oil, Amazon is ... obsessively keeping margins down for the sake of keeping margins down. Who would have to fall by the wayside for them to "win"? WalMart?
Don't look now, but WalMart has quite the online retail presence. I'm not sure how to evaluate the advantage/disadvantage of "in store pickup" but I've used it for big bulky things, when I'd needed better control over when I took possession of the item.
Meanwhile, Amazon's effect may well be part of our increasingly deflationary economic cycles.
Internet-age monopolies are an interesting issue, although I'm not sure what you mean by 'money has a natural monopolizing effect'. Markets with always decreasing average costs are prone to natural monopoly (e.g. fixed line telecommunications).
IMHO, the internet has given rise to a different kind of monopoly: 'network' monopolies. I'm not sure if Amazon could be considered one: sellers don't face high switching costs or high costs associated with selling on multiple Amazon-like platforms at the same time. Neither do consumers. So both sides of the market are fairly price sensitive, and large positive 'same-side' network effects don't appear to be present. In Amazon's case, same-side network effects might actually be negative on both sides of the market (i.e. more consumers means higher demand and therefore higher prices for consumers, more sellers means more competition for sellers and therefore lower prices for sellers).
Android and iOS, and their respective app markets, are probably better examples of network monopolies (and possibly a 'winner-takes-all' market). These kinds of monopolies pose interesting questions for government: Should they regulate? How should they regulate? Do existing anti-trust laws cover these situations?
Wikipedia has some good material, in case you're interested in some further reading:
a) control the production of the good (or service)
b) control the interface to the consumer
Everything else are middlemen and will disappear. These guys' suppliers do a) and Amazon does b). These guys are simple middlemen who add no real/critical value and will be crushed by Amazon. Just a matter of time.
In the real world, though, opportunities arise that neither "a" nor "b" take advantage of.
One of them is noted in the article...grey area product sources. If "b" isn't taking the time to find distressed inventory, there's an opportunity for a middleman to do so, and undercut "b".
There's also some value added where the middleman has expertise or something to offer that neither 'a' nor 'b' has. That could be better marketing (product photos, how-to videos), or better support (more lenient return policy, deeper technical support, etc), bundling multiple products together, niche sales, or something else.
Generally, successful middlemen find something missing between "a" and "b" and fill the gap. Is it a war of escalation, where either "a" or "b" can react? Sure...but typically middlemen can move faster.
Or, in short, they exist for a reason, and I don't see that the internet age changed anything about that other than pace of change.
In a perfect world, yes, the producer controls the amount created and the specified price of the product, but the article mentions that producers sometimes sell to one market for more than another. If the delta between these two costs is large enough, these "simple middlemen with no value" step in and earn some money.
I don't see any reason for Amazon to crush them - Amazon is getting a cut of their profits and they effectively outsource all liability to another company.
I also don't think they're simple. They've identified a constant in our world - inequality in either income or product, and they're exploiting it for the (arguable) benefit of the consumer. What they're doing takes experience and connections (to the logistics & supply chain) and it takes intelligence and engineering (to predict supply/demand and price appropriately).
Do I think these guys are inventing a world-changing technology? No, but let's not underestimate - they are far from simple and they will likely never go away.
>>Unlike eBay, where each vendor maintains a separate listings page, Amazon tidily groups its Marketplace sellers by item, hiding away the inferior offers, to showcase the best deals up front. (In seller parlance, landing the number-one spot is called "getting the buy box.")
It's a key difference. For example we purchase shipping labels on Amazon and Ebay. On Amazon it's harder to know what you're getting unless you carefully check the marketplace seller you're buying from. The intuitive process of just reordering what worked for you previously from your purchase history doesn't work on Amazon as it does on Ebay. All too often, a different seller now owns the buy box with an inferior product.
Wonder how is that after tax net income. But yes there has to be a unicorn there and this is one of those exceptions. Most Amazon retailers are complaining about getting shafted.
I wonder how long it will be before Amazon starts truly transitioning towards becoming the end to end supplier/seller. It seems to me that they have strategically used third party sellers to bankroll the needed infrastructure and provide vast amounts of data to mine. Third party sellers can't stay on Amazon forever unless they have margin, and it's a near-certainty that Amazon could capitalize on that margin better than a third party reseller can.
Seems like an awful lot of trouble for 3-6% net after COGS. From that, they have to pay their employees, rents, lawyers, etc. all with the threat that on any given day there could be wild fluctuations in their revenue. Good luck to them, but let's see how long they last. Amazon, unfortunately is a zero sum game.
Are you sure you aren't confusing net and gross? The graphic in the article pretty clearly says, "3-6% net profit margin per item". I would take that to mean after accounting for expenses. Meaning they make a profit of between 2-4 million on their 70 million in sales.
If it is really the net margin it doesn't make sense to say "per item". Maybe this metric includes all the variable costs (like product acquisition cost, salaries directly linked to handling the orders and cost of delivery) but there are still fixed costs that have to be paid from that "net."
Those costs (employees, rents, lawyers, etc) are already included. You're thinking of gross margin which is indeed based just on COGS. But the number in the infographic is net margin, and therefore covers all expenses of running the business.
Be sure your uMatrix HOSTS files are up to date, I can see a `inc.com` in the packaged Peter Lowe's list, it seems this has been removed at some point.
The opportunities on Amazon are exciting. The company is still growing, continually gaining customers. As far as I can glean it has recently been pushing to make international sales easier, which is a whole 'nother pot of gold to start snatching coins from.
That $70M is clearly their total sales, not their net income. And I'm pretty certain that when Amazon is acting purely as a marketplace intermediary, they won't be counting the full value of the order as revenue. So Amazon is making something like $7-10M a year from them.
I'm trying to figure out what their net income looks like, and I don't think it's much. They've got 16 customer service reps, over 100 packers, who knows how many other employees, refunds and replacements to deal with, plus the warehouse lease that all have to come out of their margin. What they're doing is still impressive, but they're not taking home Apple-level profits any time soon.
I know this gauche on HN, but... I'm on a macbook air. Between the "skip ad" popup and the top bar that eats the top 2/3 of my screen, the article is unreadable. After some scrolling around the top bar seems to have gone away, but I've already lost interest.
Please stop making your articles unreadable. It's not like I've disabled javascript. Actually... I suspect it would be more readable that way. Hmmmm.
But once these marketplace sellers came into being, the qualify of the experience is on the whole, lacking.
Years ago, my family would use a particular salad dressing enough that when we found it on Amazon as a "send me a case of this every three months" thing, we were happy with that. It was marginally cheaper than buying at the store, but it was far more convenient
At some point, we started receiving things from sellers that were obviously not Amazon themselves, and then endless emails asking us to rate their service, etc. Sometimes the product would be poorly package or damaged in shipping. Once it seemed like something had maybe turfed off their pantry shelf into a box to send us. It was annoying but not worth our time to 'fix', so we just canceled that order.
But lately it seems like things have been getting harder to deal with when these third party sellers are involved. I don't want to save $1.37 on a $200 router that I receive in obviously opened-and-returned condition and missing parts, and then receive an insulting letter from the (actual) seller about it somehow being my fault and saying that I should pay the return cost.
It may be possible go get amazon involved in resolving this - BUT IT IS NOT WORTH MY TIME. At some point, it was easier and cheaper to use Amazon for all these types of purchases, but Amazon Marketplace is hurting their brand. Weakest link and all that.
(Maybe we'll get a return to brick & mortar stores for radio-shack or similar things.)