Amazon to close all of its physical bookstores and '4-star' shops
reuters.com
reuters.com
They helped me realize that what I value about going to a good independent bookstore is curation. Not "highest rated" or "most popular" - but books I wouldn't get otherwise exposed to that an employee has selected to carry, and display.
That and local-interest connections to the place the bookshop is located.
Amazon's bookstores looked very much like a good independent bookstore (well, if you take away all the Ring Doorbells and stuff), but lacked that essential ingredient. Everything was just based on some "best seller" list for a specific category.
If I already know what I want, Amazon is a really easy place to get it. If I'm casually browsing, both Amazon.com and Amazon's physical bookstores were really bad at helping me discover something new that I'm interested in.
I'm physically incapable of visiting a good independent bookstore without buying something. Often far too much of something. It's an expensive disorder I have. I think the last time I walked into Elliott Bay Books with no intention of buying anything, I walked out about $300 poorer.
I've probably stopped in to Amazon's bookstore at University Village in Seattle about a dozen times since it opened. I only ever bought something once. A charging cable.
We should form a support club or something, I'm the same as you even though I live half-way around the globe. Something tells me we're not the only ones.
I'll join the club.
In Japanese, there's a word for it: tsundoku. It describes a person who buys books and mostly doesn't read them.
Often they pile up on the floor, on shelves, and assorted pieces of furniture.
Virtually my Audible resembles this.
The same applies to music. Spotify is missing that certain ingredient that local radios (before they were sold and cannibalized by Clear Channel/iHeartMedia) and early 1980s MTV really nailed. We want a knowledgeable human to guide us through the landscape of books, music, film, and everything else.
> Not "highest rated" or "most popular" - but books I wouldn't get otherwise exposed to that an employee has selected carry, and display.
In the past few years Netflix introduced their "Top 10" feature which is prominent near the top of their app. This was the reason I unsubscribed from Netflix. I had been a member since 2004. The "Top 10" feature reminded me on a daily basis that I have no interest in the content Netflix has. Previously I had assumed (because of their rather horrible discovery mechanisms) that Netflix had a much deeper catalog and I could find something to watch if I keep searching. The top 10 list made me finally realize their offerings simply were bad. The value of the service wasn't there.
Only 10 months old and trying to integrate genre now…
I don't know how you managed to get such a large list of bookshelves already, but I'm impressed.
Ya, I worked really hard in the early days to get a ton of lists and now I've built out a small PT team that helps me contact authors and shepherd them through the publishing process. We are adding ~400 lists a month, and it takes a TON of time to set it all up, vet it for quality, format, setup all the data, etc.
I've got some really great stuff coming, and I hope to leave beta toward this summer or fall.
Apart from thinking that Minecraft was set in Washington D.C., it gave multiple good-seeming suggestions for all the random topics I thought to put into it. That's pretty impressive, especially given (or, perhaps, because) your data is manually curated. Have you considered supporting Bookwyrm-flavoured ActivityPub, for (potentially curated) reviews?
BISAC has 1 to 5 categories assigned to a book, pretty standard.
THEMA is similar, but has a qualifier which is a cool concept. So you can add a location, age group, or other qualifiers to help define who the books is for.
Ya, right now we use NLP to analyze a bunch of data to to determine what the book is about. It works pretty well but I am working to improve some of that :).
"Have you considered supporting Bookwyrm-flavoured ActivityPub, for (potentially curated) reviews?"
I am not familiar, can you tell me more or send me a URL? I don't know what that is :) (ben@shepherd.com I added it to my list to check out via google search as well)
I am trying now to bring genres and other data in, my hope being that if you are on the WW2 page you can filter it to show only history, or only historical fiction, or even science fiction.
And, I would like to do similar on genre pages (yet to be built), so that if you are on a fantasy genre page, you can filter down to elves, dragons, or things like that.
Big technical challenges but I am going to see what I can do. This is something I've been wanting for a long time, and playing with some cool formats other than the current one to help people bump into awesome books/authors.
This was orignaly Pandora's big selling point.
Every track sent to them was analyzed by someone with a degree in music to help build a recommendation engine. It was a combination of algorithms and people.
I am not sure how well that scaled or if they still do that, for the longest time Pandora had a tiny selection of music and I presume the reason was their intake process.
Netflix today also has weird gaps, like if a show had 5 seasons, Netflix only has seasons 2 and 3 available.
It also didn’t hurt that the dvds had everything
With DVDs, it cost the same to send A or B or Z, and they made more money if you rented more or got a more expensive subscription that let you get more. So, they were highly incentivized to have a huge catalog and send you stuff you actually wanted.
With streaming, they want to push certain content based on what they're paid by their content partners, or what SEO tells them you will accept, and they want to do it for the lowest cost possible.
This is interesting to me, as someone who's played with Spotify and Tidal but mostly stuck with Apple Music. In part I think this is inertia -- if you stick with any of these streaming services for more than, say, half a year, then it's probably not worth switching unless there's a really compelling reason -- but all three of the services have both human-curated playlists and algorithmically generated ones, and it seems to me that Apple Music has the best curated playlists of any of those three. Spotify has the best algorithmically generated playlists of all three, but I find the best discovery trick for me is to let Apple's algorithms recommend curated playlists. :)
(Tidal, at least when I was last a regular user, was genuinely bad at algorithmically generated playlists and pretty mediocre with their curated ones. I'm not really sure what their unique strengths are beyond being willing to deeply integrate with virtually everyone and offering streams with MQA, which is basically a lossy hi-res format.)
Same here. I added a custom search keyword to chrome that automatically puts site:reddit.com after the %s.
It’s a bit of the best of both worlds, aggregation and app features like a solid tech company, and human magic that feels like local music store rather than top 20 radio stations
https://www.rottentomatoes.com/m/all_things_must_pass_the_ri...
Elliott Bay Books exemplifies that meme better than almost any place in the world. Not only does it do an almost preternatural job of putting books you might like in front of your face, but it's a literal warren that can be tricky to escape from after you've finally woken from your book-induced reverie.
That said, the whole reason this works as a store design is that the store has no particular motivation to have a big back catalog. If they don't have it, you're just gonna go get it on Amazon, so they haven't lost a sale anyway.
The store did get worse over time, though. More and more of the floor space became focused on selling a bunch of stuff that worked with Alexa. Stop it. You're not an Apple Store. Just be a book store. And as much as the book selection was pretty great to me the first few times I went in, when I stopped in a year later and the book selections were mostly the same, it was a lot less great.
Like, say I'm interested in 14th-15th century European history, maybe I have a couple good books already that I know. You could go to a university library and look them up on the shelves, and find a wealth of related books near them. Amazon won't even give you that level of browsing by library classification, using data which already exists. They have the same half-assed recommendation system which seems to have barely changed in the past 15-20 years.
It looks like there's much less focus on Amazon lists these days. Maybe they got rid of it entirely--it's hard to tell, all I can find are wish lists and such now.
No, they don't have exactly the book I've been looking for. But occasionally they have absolute gems that I'd never have found independently.
Whereas you go to the indie store and there's an entire shelf in the back dedicated to heraldry, seemingly for no other reason than that it's an area of acute personal interest for the proprietor.
Over time, their selection became poorer and poorer. Say, the "computer science" section of old, eventually lost anything like a text book, and became a few shelves of certification guides and "X for Dummies" books. Getting bought by Indigo only brought the coup de grace, when half of _those_ books were now replaced with candles and kitsch.
I actually burned myself out on books a long time ago. Very few deliver what they seem to offer on the shelf or website. Papers (individual publications curated via peer review) offer more for technical topics, and there is a lot of fiction online that is way more enticing (and…those that don’t work can be dropped after a few pages without being any poorer).
I think coffee shops demonstrate this in a way, but they don't go far enough, compared to what I'm thinking of. I imagine somewhere that has comfortable places to sit, read, and even some food, but primarily what you're paying for would be like a monthly membership.
When I travel, for example, I prefer now to stay in hostels over airbnb's, just because I like having an easy way to hangout and meet other people. I can still be online and do work, but at certain hours there's also more social activity and I appreciate both aspects.
More of everything is not necessarily better. More of everything filtered algorithmically can be actively worse - not just because of poor modelling and information loss, but because the goals of the algorithm (typically related to some combination of hard sell and addiction) are diametrically opposed to a good human experience.
There is such a thing as culture, and algos are incredibly toxic to it.
Incidentally your semi-social idea is exactly what "clubs" provide in the UK.
Not many people know about clubs, which is strange, because they're incredibly influential on politics and culture.
They're subscription-only spaces for mostly upper middle class people in an expensive building - most in central London, a few scattered around other big cities - that offer food, social contact, and networking opportunities.
Most have curated selective membership. You have to be invited and approved before you can join.
They're also split by profession - politicians, lawyers, media personalities, musicians, journalists, creatives, and so on.
There used to be one for London startups. (Possibly still is - I haven't checked recently.)
When I was growing up book stores were kind of that place, but most of that has gone away.
This is something I am bothered by: if top ten lists are based on what people buy, and what people buy is based on top ten lists, where does humanity enter into the picture?
In the future we will have advanced AI like GPT-3 trained on the internet, generating content for the internet at far larger scales than humans, with that content being consumed to train the next generation of AI, like a snake eating its own tail and pushing human content into irrelevance. I feel like there’s the potential here for a dystopian novel where future humans are desperately trying to connect with another human in a giant maze of algorithmically generated conversations served up by AI bots through the metaverse.
What I knew: these particular people had left for this project. The head of it was the guy who had lead the Kindle project- he had rejoined Amazon allegedly because Jeff asked him to lead this project. And for whatever reason, this JIHM org had all these new physical book store managers reporting up to them.
Over lunch one day, I proposed a theory to my coworker: "they're going to make that IBM commercial from the 90s, where the sketchy dude grabs a bunch of stuff and puts it in his pockets, then just walks out the door and gets a receipt". It was a fun idea, and given the CV people they had pulled it may have even made sense. "What about the book stores?" My friend asked "It's a smoke screen. That same guy leading JIHM is legendary for leaking false reports that Amazon was going to build a PC to cover him hiring hardware engineer for the Kindle."
6 months later, Amazon Go with it's 'Just Walk Out' technology was publicly announced. The CV guy I knew was even in the commercial ("Holy shit, is that Danny?!").
Maybe this is the final conclusion that those stores were just a cover.
(Caveat: this is all wild speculation, rumours, and probably bad memory on my part, but it's a great story and I swear it's all true!)
Might be wrong, this is just my observation in SEA/what I recall.
On the other hand, maybe that's the real lesson of Grocery Gateway and "dot coms" who tried to do a hard burn on this back in the day— it's a really hard thing to get right, and getting it right requires an extremely long horizon of slow and steady growth.
The goal of Amazon fresh was to say "hey UPS! Looks like we can do deliveries all by ourselves without you! Give us a better deal or we'll become your competitor". Because all those deliveries were on Amazon trucks.
Then one day Amazon's delivery needs scaled beyond what UPS was willing to take on. And so Amazon Logistics became a major thing.
Resistance is futile.
https://www.nytimes.com/2022/02/28/technology/whole-foods-am...
(Sorry for paywalled article)
From what I've read, today it's so expensive in terms of hardware that it's probably a money pit. But hardware gets cheaper, software gets more efficient, and eventually this stuff will just be in every store.
(not true)
https://www.bloomberg.com/news/features/2019-07-18/amazon-s-...
The biggest "risk" for my use has been that Ocado's predictive pre-filling of my weekly order is just good enough that I've gotten complacent about checking the orders, but it's not quite perfect, so if I miss the deadline, I might occasionally get stuff I didn't need (this is optional functionality - you can just place orders when you need them too and you can tweak what is included to have more control).
Ocado, the company in question is licensing various parts of their tech stack. E.g. in the US they've licensed to Kroger (which also took a 5% stake): https://www.businessinsider.com/ocado-kroger-amazon-online-g...
Between that, and the option of (more expensive) rapid deliveries from takeaway delivery companies like Deliveroo, or from the multiple different 15-minute grocery delivery companies, grocery stores just seems like a nuisance to me at this point. Of course my judgement is influenced by the fact that I don't have a car so not needing to drag shopping on foot or bus makes the leap in convenience far more significant than for a lot of people.
It ended up being Kindle ads for Amazon Local, the now defunct Groupon clone. So anti-climactic after all the cloak and daggers theatrics. At least Amazon Go is a much bigger deal with interesting tech behind it.
So it might have been a doomed concept from the start, but I'll miss having a physical showcase for cheap knock-off and import goods.
Part of the appeal of a bookstore is to have a different shopping experience than what you get online or even at other physical stores. It's walking over to a section, picking up a book, seeing what resonates. It's 'browsing' in a slower, calmer form, closer to the reading experience itself.
Why doesn't Amazon try to make that? Replicate that experience in a new brand, call it 'booklovers.com' or something. Make it a cut down subset of features and information, with the book's content more at the center of the experience and easier to browse, along with the books 'next to' it. More curated, less metadata.
Being freed of physical world means we can have many very different experiences without dealing with the scarcity of real estate. Yet Amazon tries to standardize the shopping experience, to the detriment of many categories of products that deserve their own approach.
Amazon makes so much money from bullshit at retail that they cannot enable basic features that even not so great physical retailers get right. Case in point: try buying windshield wipers for multiple cars. You’ll get ads for incompatible blades and tricked into a subscription. It’s literally worse than the 1980 experience of flipping through a torn, filthy book hanging from a shelf by a string.
Why don't they also use AI to make highly pleasing storefronts and target other markets?
Amazon is the everything store, but it presents like a big box store. Why not build mini Boutique stores or more of an IKEA experience?
Another funny one is lightbulbs - good luck unless you are looking for a particular part number.
What OP is proposing is the vision to divorce the URL from the warehouses. Allow some teams within Amazon to make websites to buy books or other stuff from bespoke URLs. I think at the scale that Amazon has, it would only help.
I use NLP on the backend to tie books to Wikipedia topics and working to improve that and bring in genre filters/pages soon.
Let me know what you think (ben@shepherd.com)
Now that I think of it.. they do experiment with things like this. For example, near the top of my Amazon homepage right now, I see "Influencer fashion finds". Clicking in reveals some kind of modernized QVC social media livestream experience.
Further down on the homepage is a combination of what appears to be algorithmically curated shopping lists, and human curated lists.
It's all very messy and disorienting to navigate, but I guess they're.. trying?
It will be interesting to see how their grocery and convenience stores fare. I have some Amazon Go stores nearby and they never seem too busy.
As always, I'm sure the explanation for why their search engine and recommendations are so bad is either that they don't think it would make them enough more money to be worth it to improve them. An even more sinister thought would be that they think it would make them less money to give consumers better recommendations. I'm not sure which it is, TBH.
Counterintuitively, you're wrong about this!
Some astronomical number of Amazon purchases are returned. Like 10 or 15%. So anyone who has just bought a TV is extremely likely (compared to the ambient) to be buying a TV right now.
It's algorithms all the way down.
If you're not returning the TV, this useless recommendation trains you to ignore the suggestions altogether. This must have some cost, but it's not one the algorithm can directly track, so it never gets priced into the recommendations.
But when I do consider returning an item due to dissatisfaction (not DOA or damage), I definitely research other options before initiating the return.
This much makes sense to me.
I suppose there's a possibility that your ML algorithm could decide that the profit from selling a TV so greatly exceeds that of the accessories that a 10-15% probability of preserving the sale in case of a return might be more worth it, but I don't really know the ins and outs of retail ecommerce to make a judgement there. It still just seems silly to me. ¯\_(ツ)_/¯
I'd be blown away with surprise to learn that AMZN hasn't optimized for revenue in the calculation though, so it seems likely that we're wrong.
Of course we'd prefer AMZN to optimize for customer happiness/benefit instead. To me, the "smart" approach would be to recommend accessories until and unless the customer shows some signal indicating that they are looking for another TV. Then switch to full replacement strategy (or even return-minimization strategy) mode.
That or similar logical approach may have been abandoned after getting better results (by some metric) from the ML-optimized approach. Assuming there's a human involved at all.
So much easier than hunting through a Barnes and Noble or doom-scrolling through online listings
The biggest difference to Amazon is after I pick that I want an HDMI cable, a garden trowel and a high temperature incandescent bulb for an oven, it says ready to pick up in 5 minutes and I can walk to the store (five minutes) and get it with no delivery surcharge. Also I could pay cash (but I don't).
Obviously if you live 2+ hours from an Argos, Amazon is starting to look more competitive, but in the UK who lives two hours from an Argos? I don't buy much from Amazon.
definitely a gimmicky thing to put on the receipt, but I do miss these little "fun" things
Guilty as charged on consumerism front, but I do exactly this on black fridays.
Much less liability for the retailer, and puts the onus for sales on the manufacturer. I would be surprised if Amazon didn't try to do that with their 4-star shops.
A store that lacks any direction, curation, and empathy is more like the front end or ad space than the body of the store or programming. It is not that they are focused on what they want to sell you of course they are. It's that they have put less thought in what they would want to buy if they put themselves in your place. An algorithm is a shit way to stock a store.
1. Even Amazon's own bricks-and-mortar stores find it impossible to compete against Amazon's online store.
Or:
2. The physical stores were just a "front" necessary for developing and testing new technology (as suggested by mabbo elsewhere on this page https://news.ycombinator.com/item?id=30533453).
My money is on #2.
RIP the Powell's technical book store. It was a very large book store with only technical books. I spent many a happy hour poking around and finding interesting books.
It was pretty weird.
As others said, it was covers out, which in my opinion only had the downside of having limited shelf space.
I've always found the physical similarity of Amazon's own retail store efforts to Waterstone's curious, as Amazon actually tried a partnership with Waterstones to sell Kindles and such years before opening their first stores back in 2012... First Amazon book store with remarkably similar design and layout to a Waterstones followed in 2015. As a UK expat living in the US my first thoughts on entering an Amazon Book store here was just how much it was like Waterstones.
Could just be coincidence!
> https://www.independent.co.uk/news/business/news/waterstones...
> https://www.theverge.com/2015/11/2/9661556/amazon-books-firs...
Someone probably realized that Amazon Go & Whole Foods were better for foot traffic (since they sell groceries & other items) and investing more into selling Amazon Go walk out tech would be a better expansion plan than their own stores.
I didn't really see the point, given that some places like Wal-Mart or Best Buy already have display models you can play with.
If you're wondering how it detects the produce, produce in the US has tiny little stickers on it or at least a PLU code where its sitting. You enter that 4 digit code and it gets weighed.
Felt like a "scaled-up" convenience store. Not having a line was very nice.
The 4-star store out at Oakbrook Mall though — I didn’t find as interesting but I’m not in the demographic that shops for gifts for others. It was pretty busy though.
Amazon will close its 4-star, pop up and bookstore locations on various dates and *notify customers via signage*.
Remember thinking at the time that it was odd, but makes way more sense now.
I wonder how long they'll keep the Amazon Fresh brick-and-mortar experiment running.
It was a bit like a Sharper Image but instead of 80% of the items being useless gadgets it was more like 20%.
The foot traffic seemed to be about people returning stuff and then converting to customers of random stuff.
It was an awesome experience, and I'm sad that this might be the last time I get immediate service like that from Amazon.
I don't see why I'd want to go to an Amazon Shop instead of Wal-mart, Canadian Tire, Target, whatever else already exists. It's a market with razor-thin margins that's absolutely doesn't fit their current approach and growth focus.
I realize they meant it as "everything 4 stars and above", but . . .
https://www.businessinsider.com/workers-first-union-push-at-...
I've been noticing this more and more lately, and it makes me curious as to whether there's any research into the psychology behind contradictions.
Beyond that, I've only ever used them as a quick way to return an item to the warehouse.
After reading through some of the comments here, apparently the thing that I went to was some kind of Amazon mall "kiosk", not a proper store. Whatever it was that I went to, it was just incredibly creepy. It was just a bunch of kindles and Alexa's crammed into this little area, with no one around, hidden away in a weird area of the mall.
I imagine it was supposed to have "go" associated to it in some way? There was no information I saw anywhere indicating how any of it was supposed to work. I didn't any sort of cash register or obvious spot where an employee would be. The whole thing was really just gross. It felt like, if you were to see a purse in the middle of the street with fists full of dollar bills sticking out of it with no one around. It's like someone put it there, specifically with the intention of them wanting you to take it. It's like it's accusing you.
It was this disgusting sensation of being monitored in a way that was outright nauseating. I can't really describe it. If you've ever gone to a gas station or some other place, and are trying to pay for something, and the cashier comes out 5 to 10 minutes later from the bathroom, it was kind of like that, but without the relief of seeing a human after a few minutes. Now picture that, but in a small square space without walls, or signs of life. It's like no one's been there in months. That's what that location felt like.
Anyway, if these Amazon stores were anything like that, I'm hardly surprised they didn't do so well. I'm not a huge fan of the idea of being treated like a criminal, just by entering a store.
The names sound like they could have used some serious rethinking, too. My first assumption after hearing the term Amazon "4-star" is that it's not supposed to be very good. 5 stars is good, so if it's only 4 it's low quality products only. Was it meant to be a "high end" dollar store? If so, why wouldn't I just go to the dollar store instead?
I'm sure they have their reasons, I'm not vain enough to think I'm smarter than the people who made these decisions, but I can't imagine why they bothered, or if in doing so, why they made such weird decisions.
Seems like a way to show off the AR capabilities Amazon is working on. Secondarily, many haircare products are only sold in salons and this effort could eventually be a way around those restrictions.
They were a convenient way to return incorrect products, though.
If someone is willing to pay with a card, especially if they're worried about COVID contamination, why would they go to a store for something they can buy at Amazon in the first place? The reports seems to pretty consistently say that surface contamination is a much less likely vector for spread than human proximity, and going to a store still involves human proximity.