And where is the competition? Is there anyone at all who can provide something like this?
There are competitive options, if you’re in retail you do not have to give Amazon all your “retail data” or use their tech.
Conclusion : ideas are worthless, Execution is key
This was also often the conclusion found during the Dark Ages. ;D
Do you mean to say that doing something well is way better than imagining doing something well?
Ie. Good ideas are valuable
To me, that's part of the reason why thought pieces are valuable to me. Also, unconventional tracks at academic conferences, like provocations at DIS 2020 [1] (which I cite here since it's on my mind because I'm personally applying to it right now).
Execution can bring profit. Ideas might not.
History is full of examples of visionaries coming up with ideas that have inspired and fueled others--in some cases whole industries have been founded on these ideas. For instance, Doug Engelbart gave the "Mother of All Demos" in 1968 [1]. It was a proof of concept, but the ideas he talked about were implemented by other companies--who profited from it.
It is unfortunate that we live in a world where ideas hold such little monetary value, but they're most certainly not worthless.
But let's say you have very good ideas but you're a mediocre executioner.
1. Could you find a very lucrative job/partnership mostly because of your ideas, and get quality help with the execution ? It's possible.
2. Ideas are easy to steal. Do you have a decent strategy to prevent that from happening ? Those sometimes exist.
3. Not every business sucseeds. Can you try repeatedly ? Maybe, depending on context.
https://en.wikipedia.org/wiki/You_Will
I was always confused as to why they'd spend money on vaporware advertising, since I don't think they were responsible for doing any of the things.
I get why they spent money on vaporware advertising.
But I don't get why there's no modern equivalent. No one seems to be making equivalent statements about technology that's 20-30 years away from today.
There are vague hints that AI and QC will somehow be awesome, but they're very light on specifics.
It's interesting how we got from the forward-looking 90s to the backward-looking 20s.
There are some fully automated stores the size of shipping containers in Beijing, but they're mostly demos, too.
I don't think anybody really has this working well enough to deploy. It has that machine learning "we got to 95%" look. This is something that's easy to almost do, and the last 5% is really tough.
I do love how the very first banner video standard.ai shows is how one person passes an item to another. I'm curious to see all the edge cases Amazon Go ran into over the year or two they ran. I can see a lot of messy cases like that coming up.
Standard.ai apparently has a demo store in SF, although it's been closed due to the coronavirus epidemic.
Grabango - one installation in test at a Giant Eagle store.
Getzippin - one installation in test at a Lojas Americanas store. Their site gives the impression that it's really about getting people to install their phone app, so they can be spied upon.
v7labs - we're AI, we don't need installations.
It's always been a question of when the supporting technology was going to be good enough to make it work. The idea has been around long enough for most of the early patents to expire.
Yep, I meant more along the lines of how you put it i.e. packaged up and ready to start selling it as a boxed-up/drop-in solution.
All the way down in Australia.
At PIXEVIA we develop the technology and run a fully working convenience store in Europe.
Although not yet doing automatic checkout, but most of the technology needed for it is there.
What if I grab two items with one hand, and then put one back? Will the system correctly track this?
In hindsight it seems obvious that the play was to "AWS" the whole thing: make yourself the first/best customer, nail the implementation, then sell it to the world.
Amazon were slightly profitable or break-even since 2003. There's a convenient chart of their profits since day one here:
https://qz.com/1196256/it-took-amazon-amzn-14-years-to-make-...
Starting in 1997 they bled money with mounting losses until about 1999, when they began to turn things around. The dotcom pop is clearly visible but they recovered almost immediately and their losses continued to shrink until about 2001-2002 when they became break even. From 2002-2011 they either made small profits or nothing, but that was obviously because they were growing at a rapid pace and putting all the money back into the business. Once AWS launches in 2006 (so about 10 years after day 1 in retail) profits start growing but then are back into the red around the time of the GFC+recession, and again in the 2012-2013 European recession. After that it's stratospheric profits.
How much investor money is "infinite money"? Somewhere between $8-$9 million before they floated on the stock market.
https://www.quora.com/Who-were-the-original-investors-in-Ama...
Obviously, investors who put money into their IPO have done extremely well and cannot claim they were shovelling money into a furnace, far from it.
The inflation in investment round sizes over the past 20 years has been staggering. I see nothing that suggests Amazon was unusual in raising so little money (comparatively speaking) before they went public.
Finally, Amazon's marketplace is just not navigable. Their search is pathetically bad, and this is nothing new, it's been a common complaint for years. Some fairly massive amount of their traffic is inbound from Google searches like "some product amazon" just because of how pathetically bad it is. The only other reasonable way to navigate their site is if their similar product suggestions or "commonly bought together" happens to nail the item you were looking for.
I've had searches where adding an additional keyword that is in the product title will actually cause the product to disappear from the search. What in the actual fuck.
That's what I thought until I was scammed by a Chinese on eBay... after countless emails, calls, even police reports, eBay did not return the 800 dollars I lost... not sure if it's an isolated case but it was pretty frustrating.
Their recommendation engine for related prducts used to be nice but that's been replaced by sponsored products which lowers the quality.
Now, the only way I use Amazon's search is when I have a SKU or part number and type it in directly. But I also do that in Google and often find the same thing elsewhere at similar or less cost (including shipping, though to be fair it may take a day or 2 longer).
eBay's search is still one of the best for me, in that it respects the keywords I put in and allows for extended query syntax to really hunt something down (which then can be turned into a saved search).
One of the most creative things I found on eBay was tool rental. I needed a tool to replace the bearings on my washing machine, and a seller was selling one explicitely as a rental: tool was charged about $120, with $35 shipping. When done, sent it back for a refund. The "shipping" included the rental fee and shipping both ways.
The one that they're missing that I really wish they would include is "multi item BIN" formats. People will list a $1 item so that they're the first result that comes up, and then the item you're looking for is high priced. The prices are in fact so high that I go out of my way to try and exclude these items using "not" keywords, setting a minimum price, and filtering to US only, which collectively get most of them.
I have a lot of saved searches for "rare" items that only get listed infrequently, or for items that I'm waiting to come down in price.
The contrast between the way those two sites handle their search is stark.
After shipping with Amazon for our non-Amazon orders for six months, we've moved about 80% of our shipments to them because they save us that much time, hassle, and money. FedEx picks up almost all the rest, with a handful of packages going out UPS and Postal due to address requirements, or customer preference on phone orders (all online orders are Amazon/eBay/website where we limit shipping choices; we are about 60% Amazon, 30% eBay, 8% website, and 2% phone orders on a typical day).
I hope that there's always a USPS option. My apartment complex has package lockers that can only be used by the USPS, so that's the only safe way to get packages delivered when I'm not home.
If they released this at Whole Foods, I'd guess they'd fire all the cashiers there. That would make this rollout a lot less exciting for a lot of people.
I hope Amazon comes up with an awesome way to help the cashiers this technology replaces. Then if self-driving cars end up working out, that industry can learn from what Amazon did.
[0] https://www.cnsnews.com/news/article/rudy-takala/top-2-us-jo...
Since Whole Foods sells alcohol, I doubt that. There are things that lots of governments won't let you have customers “just walk out” with.
As I wrote that I realized they could even employ cashiers to do that instead.
They check ID entering the liquor section (and it is a human that checks), but otherwise it is bag it and just walk out.
They can be retrained to work in fulfillment centers. Amazon will provide the piss bottles.
I would think the majority would be employed at little restaurants, small shops, etc.
> Will people still be working in stores with Just Walk Out technology?
> Yes. Retailers will still employ store associates to greet and answer shoppers' questions, stock the shelves, check IDs for the purchasing of certain goods, and more - their roles have simply shifted to focus on more valuable activities.
The benefit is a much better customer experience, not labor savings.
The primary benefit is access to a lot better data about shoppers, not a better customer experience.
It's a massive competitive advantage far beyond whatever data they get by watching me roam the store.
Retailers that compete on "experience" like Target etc will reinvest into a better customer experience.
Retailers that compete on price like Aldis will focus solely on driving cost down.
I'm not sure why this should be on Amazon. It's on all of us to better train people to have the jobs of the future. Our education system hasn't really changed in 100+ years. It needs to.
And if these companies profit, they do so as a reward for spending the capital (dollars and manhours) in creating new technologies that improve productivity. Specifically total factor productivity in https://en.m.wikipedia.org/wiki/Cobb%E2%80%93Douglas_product...
It's on all of us because it really isn't on anyone in particular, surely not on Amazon.
I can take your statement and flip it around just as easily: why should Amazon bear the costs of paying for many R&D ideas that may or may not pay off? We should all share that cost if we're all going to benefit from those innovations
Are you seriously trying to suggest everyone benefits from Amazon’s innovations?
I'm not saying it's a great thing, just that the jobs Just Walk Out will be replacing are already being replaced by self checkout.
Recently came back to the East Bay (SF) and the Whole Foods I'm staying near, more than the others, seems like a staging store for Buy Online & 1) Get Delivered, 2) Pick Up In Store.
There are often more WF employees shopping for pick-up orders than customers walking around the store. Navigating aisles that are full of these WF shoppers gives the store a pretty signficantly different feel than the other stores that I'm a regular at. Plus, the staging/storage area at the front for ready-orders is much bigger than at other stores.
Not as true at SF city WFs, or the ones in Reno, LA, Park City and Phoenix/Scottsdale that I'm familiar with.
I'm guessing this dynamic, plus "Just walk out" is where they're headed, at least in certain densities. And that's where some of the cashiers may be transitioned.
It's not just about "AWS"-ing the thing. It's about gaining access to all of the transaction data and marrying that to what Amazon already knows about you. A retailer like Walmart would NEVER use this tech, they know their data is one of the most valuable things they have (in fact they don't share their data with anyone, while all the other FDM retailers do), but boutique retailers like J Crew or whoever would happily.
Amazon is always putting up ads for me based on my purchase history, but they're never what I wind up buying. I have no explanation for what is wrong with their algorithm.
Just call me The Hammock Man!
This was about the same time they stopped shipping things in cardboard boxes and went with bubble packs, and most recently, paper wrap with no padding whatsoever.
Basically, you choose why you're returning. In some cases if you say something other than "the item was defective", it'll only have options that cost money.
It's super rare, but I've seen it before.
Other things however come just as you would expect (paper wrapping or bubble wrapped). So it all depends on the amazon worker doing his or her job..
It's pretty obvious that people don't go around shopping for screen protectors and then buy phones to match.
But I wonder why they don't just flip the prediction? If you bought X, then advertise anything but X.
Like if you bought a toilet seat, they might advertise toilet paper instead (which would be good, because all the stores around here just ran out...)
Or maybe showing a wild selection of weird products results in people "brainstorm" other shopping needs, as a kind of anchoring.
https://en.wikipedia.org/wiki/Anchoring#Anchoring_in_negotia...
https://www.forbes.com/sites/kashmirhill/2012/02/16/how-targ...
I wonder if their creepy surveillance tech knows I said it. I was going there to pick up my brother's medicine from the pharmacy. It's a real shame. I used to LOVE shopping at Target. But the dang store has gotten greedier and super obnoxious over the years.
I was positive they used to use their tracking data to design store layouts with minimal clashing, such that they distribute the shoppers across the store. It seemed to really work, too.
In their newer layouts, I dunno what it is, but everything seems crammed together all in the same spot. They replaced the popcorn, icees, and hot dogs, with food I don't like that costs too much. They got rid of the handbasket things that were handy and all over the store, and got rid of the benches too. Then they sold their pharmacy to CVS.. which immediately threw out the kickass containers Target used to use and added a super obnoxious phone system.
Just overall, I hate shopping at Target now.
Turns out people who have just purchased something are far more likely to purchase that thing again than a random person.
This makes them the perfect person to advertise to.
Even if the percentage is quite low this can be better than advertising to the general population. Say the numbers were 1 in 10 000 impressions leads to a toilet seat sale, but 1 in 1000 people who have just bought a toilet seat buy another one soon after (first one was broken/distributor/toilet seat collector/remodelling whole home/etc).
If those were the numbers it would be silly to not advertise to recent buyers.
Essentially, the self selection of 'I just bought <x>' very often puts you in a category that is far more likely to buy <x> again than almost anyone else.
Very helpful for identifying what’s hot and releasing your own version. Figuring out seasonal trends. Or getting competitive price information to match/beet. Walmart brags about their ability to presciently stock stores, which keeps turnover & capital efficiency high.
It looked basically the same, except that there were these huge, bright blue Amazon Prime ads everywhere. It felt vaguely dystopian—a store of my youth invaded by the giant tech monolith.
If Amazon is trying to not noticeably change Whole Foods, they're doing a pretty lousy job.
WF has already served Amazon well by being a test bed for grocery delivery optimization, no need to screw up a profitable existing business with any additional radical changes. That's what Amazon Go stores are for, and now they can sell that tech to other stores. Once the tech is mass-adopted, they can smoothly switch all WF stores to "just walk out" tech with very little complaints, as that's the experience people would be already used to at all the other stores
Not only that, Amazon would be literally paid by others to take over Amazon's own risk. This is such a nice play all around, they just need to manage not screwing it up somehow. But execution never was a weakness for Amazon imo, so I bet it will mostly go smoothly.
If I was a major business, I wouldn't pay Amazon to be a guinea pig, especially when I see them unwilling to do it at their own major retail chain.
If you've got a Whole Foods and a less-upmarket grocery store equally far from you, you're gonna get your staple foods from the less-upmarket store, and just go to the Whole Foods for the things only they carry.
But if you're making an online order with a fixed-overhead delivery fee from Whole Foods, you're not going to make a separate online order with a fixed-overhead delivery fee from the less-upmarket store; so you're going to end up ordering the staple foods from Whole Foods, too.
If I were doing this, I would be far more likely to make a separate order online than physically. It takes no effort to do it online, but to shop at more than one physical store is additional inconvenience.
I'm guessing they're slowly scaling up.
Or, just over charge by a large enough margin to make up for it.
For example, I made two apple pies the other day. The recipe calls for 2 lbs of apples each, so I bought 5 lbs of apples (to account for the weight loss from coring and peeling).
Based on the video below, it would seem there are a few bugs to work out. For example, never pick up a whisk at an an Amazon Go Grocery store unless you intend to buy it. You will always be charged. [1]
> How does it handle fruits by weight
Fruit aren't priced by weight. This is also covered in the same video.
So far Amazon has also not shown their technology deployed in an existing store, rather than an Amazon Go. We're actively working on being the first to demonstrate this. Stay tuned over the next couple of months.
But, sincerely, good luck. New technology should shake up the existing players, not be an occasion for oligopolies to become even more powerful.
Can you give us a sense of how many transactions standard is doing every day with it's tech that's further along / more deployed than Amazon? At least the SF location I visited had steady traffic so Amazon is collecting some real data (selection a bit limited so go grocery interesting to me)
One of the big tech companies in the 80's demonstrated this. I think it was either IBM or AT&T/Bellcore. Essentially, it was just a scanner at the door that read RFID tags in each item, and your credit card.
This was back when RFID was still oh-holy-shit technology.
The reason I think AT&T might have been involved is that the video demonstration was very similar to AT&T "You will" series of commercials that aired around the same time.
Yes, you'll have to recharge your shopping carts/baskets. Just design them to ensure they pass through charge (like a string of Christmas lights) when stacked together, so you only have to plug in (or have a cradle for) the bottom one (for baskets) or frontmost one (for carts.)
It's still a worse idea than using computer vision, if you've got that, but it's not something we couldn't have pulled off a decade or more ago.
The amazon thing is obviously based on computer vision which in the hindsight makes sense. And makes more sense than our ideas that includes CV as one of the source into sensor fusion magic...
Is this still a thing? In Poland most shops allows you to scan and pay for products in matter of seconds without any help. No lines, no hassle. :)
I'd be really impressed if any large supermarket chain (outside of the US) would be ok to be so tightly coupled to amazon for virtually no benefits compared to self checkout. It's like a lot of what SV is pumping out, half assed solution to non existing problems.
In my part of the US this is common as well, but there's almost always a line to use the scanners that is about the same as the line for the a human checkout clerk. So I almost always opt for the human.
Well, who knows how much interest will truly materialize, particularly given the data leaked to amazon, but the post claims/implies that retailers reached out to Amazon and were interested in licensing the technology.
This is an industry with a lot of inertia, and customers are not going to be driving this change like a lot of optimists at amazon might expect. They will struggle when the tech isn't working right and prefer the old ways, as they do now.
Plus there is the rampant theft that will occur, unless you hire someone to stand there and man the door. And at that point you've spent millions on this fancy tech that your shareholders salivated over to have the same exact payroll overhead as you did when you had someone sitting there scanning items.
If this makes amazon and the dime a dozen clone startups some money, good for them, but I don't see this changing the world.
Technically Walmart did this in their Rogers, AR flagship store in the late 90s, but consumers found it awkward/error prone and didn’t use it much. I have a feeling Walmart shockingly enough was just to far ahead of the curve technology and culture wise (if memory serves, self checkout wasn’t even a common thing back then or if so, was just becoming one).
(I believe M&S has something similar?)
Isn't the better question, "where is the product"? Is it a gimmick?
This looks awesome, but until I see it in action, working flawlessly, outside of the Seattle/Silicon Valley nexus, I'm not sure there will be (or needs to be) any competition.
I’ve seen a lot of very intolerable lines in the suburbs. It’s not unusual for me to wait 15 minutes to checkout at Target or Costco and last December I remember a 25 minute line.
I think this is a product. There’s about six different grocery store within a fifteen minute drive from my house. I’ll switch to the first one to implement this system.
This really doens't sound like a problem that needs amazon and "technologies used in self-driving cars: computer vision, sensor fusion, and deep learning"
Um, Walmart and most major grocery stores already have self checkout areas. So “wait in line to get served” isn’t quite the comparison. Most companies will give zero fucks about adopting this unless it becomes the expected norm.
Edit: The "scan your own cart" model is a more compelling counter-argument than the self checkout machines currently at Wal-Mart, as this can accommodate a much higher throughput.