Become a fan of Slashdot on Facebook

 


Forgot your password?
Close
typodupeerror
×
Businesses

Amazon's Most Ambitious Research Project Is a Convenience Store (bloomberg.com) 94

Posted by msmash from the what's-in-the-store dept.
Amazon has set up 14 Amazon Go stores in Chicago, New York, San Francisco, and Seattle. They do not have any cash registers so once customers have scanned a screen from a special app on their phone at the entrance, they just grab their items and walk out the door, while Amazon magically charges their credit card. By all accounts, the company intends to open more of these stores in the months and years ahead. Bloomberg Businessweek reports today the kind of investment Amazon has made into these stores -- it is the ecommerce firm's most ambitious research project to date -- but despite that, how these stores are just like 7-Eleven stores, but with more complexity and cost.

From the report: From a technological perspective, the Go stores are a marvel -- a succinct demonstration of Amazon's capacity to devote vast resources toward applying the state of the art in artificial intelligence to an everyday problem. They also illustrate the company's tendency to pursue technology for technology's sake (see: the Fire Phone), resulting in a store that offers all the selection of a 7-Eleven, but with more complexity and cost. Scores of cameras pointed at all angles hang from the ceilings to track shoppers as they wander the aisles, while precise scales embedded in the shelves tabulate products down to the gram to figure out which ones have been picked up. Behind the scenes, sophisticated image recognition algorithms decide who took what -- with Amazon workers in offices available to review footage to ensure shoppers are accurately charged. Each store also has a local staff on hand to help people download the Go app, restock shelves, and, in locations with a liquor section, check IDs.

Will all this work be worth it? Some Go stores seem almost deserted except for the lunchtime rush. Employees familiar with Amazon's internal projections say the outlets in Chicago, in particular, are falling short of expectations, and the company has had to resort to raffles and giveaways of tote bags and other branded goodies. Yet, as the turbulent history of the project suggests, the Go store isn't so much the culmination of the company's efforts but something closer to an ongoing experiment. And the potential prize -- a big piece of the $12 trillion grocery industry -- is one that Amazon, with its limitless resources and appetite for risk, may be in the best position to claim.
This discussion has been archived. No new comments can be posted.

Amazon's Most Ambitious Research Project Is a Convenience Store More

Amazon's Most Ambitious Research Project Is a Convenience Store

Comments Filter:

  • Everyday problem? (Score:3)

    by DogDude ( 805747 ) on Thursday July 18, 2019 @04:48PM (#58947964)
    What's the "everyday problem" that a $15/hour clerk can't solve, exactly? This seems like an insanely stupid waste of money and effort into solving a non-problem.

    Heck, it even makes the shopping experience significantly worse on many different levels. I can't imagine what kind of person would voluntarily go into one of these places.

    • What's the "everyday problem" that a $15/hour clerk can't solve, exactly?

      $15 / hour x 24 hours = $360
      $360 x 365 = $131,400

      In 5 years, that is $657,000.

      Add 40% for taxes, benefits, overhead, management = $919,000.

      That can pay for a lot of cameras and scales.

      If Amazon can make this work, it will pay off big time.

      • Re: (Score:3)

        by DogDude ( 805747 )
        You're right. Assuming nothing ever goes wrong. The amount of complexity in running a convenience store is extreme. Anything and everything that can go wrong, that a human could easily take care of, will never make these sorts of setups profitable. Plumbing problem? A human turns off the water and calls a plumber. Amazon store floods, everything destroyed. Electrical problem? Vandalism problem inside the store? Vandalism problem outside the store? Who's receiving deliveries? Who is cleaning up me

      • Re: (Score:2, Insightful)

        by Anonymous Coward

        Sure if all a clerk did was scan items and collect money your post would make sense and the store would profit. But you need to keep in mind all of the services a clerk provides that need to either be automated or handled on an ad-hoc basis in a store.

        Who cleans the floors? Who cleans up spills if a customer drops something? Who cleans the bathroom? Who cooks or prepares food items? Who or what discourages minor theft? Who or what restocks the shelves? Who unpacks inventory when it arrives on the truck? Wh

        • Re: (Score:1)

          by Anonymous Coward

          Who cleans the floors?
          Robots

          Who cleans up spills if a customer drops something?
          Robots

          Who cleans the bathroom?
          Robots

          Who cooks or prepares food items?
          Chef / Cooks (for now)

          Who or what discourages minor theft?
          Cameras / Anti-Theft systems

          Robots

          Who or what restocks the shelves?
          Robots

          Who unpacks inventory when it arrives on the truck?
          Robots

          Who answers the phone and provides information to potential customers?
          Amazon customer service (or third party robots)

        • Someone who's stocking the shelves, cleaning an aisle, cleaning the bathroom or preparing food cannot simultaneously be manning checkout.

          That said, Amazon's real problem and real competition is self-checkout. Millions of us routinely use self-checkout already. Ultimately, Amazon has to (a) convince customers that the convenience of not having to self-checkout is worth paying more for or (b) make their system cheaper than self-checkout, which will be quite a challenge.

          • At Aldi and Lidl, employees rotate between stocking, cleaning and manning the tills when it gets busy. Not sure how much more money this is going to save.

      • If Amazon can make this work, it will pay off big time.

        They can't. If a store is truly unmanned, it will be vandalized and looted daily in any urban center. Cameras? Please. All the major cities are moving to decriminalize theft and burglary.

    • Recognizing people and what they're doing is really a foundational capability for almost anything one might call AI.

    • "What's the "everyday problem" that a $15/hour clerk can't solve, exactly?"

      Length of queue given that the number of servers cannot be fractional.

      As human clerks are added and removed the store is still incentivized to have slightly less clerk capacity online than the potential customer flow, causing queues to lengthen. This contrasts with the customer's goal when choosing a convenience store which is to minimize shopping time. As the marginal costs of Amazon's no-checkout system are nearly zero, it is easy

    • Re: (Score:2)

      by gosand ( 234100 )

      What's the "everyday problem" that a $15/hour clerk can't solve, exactly? This seems like an insanely stupid waste of money and effort into solving a non-problem.
      Heck, it even makes the shopping experience significantly worse on many different levels. I can't imagine what kind of person would voluntarily go into one of these places.

      No cashiers (or cash) to get robbed? Seems like it would solve that problem. Shoplifting? Maybe. (or maybe not, yet). We don't know the long-term goal. Just like Prime seemed to be a stupid idea - yet LOTS of people love it. Same with self-driving cars (which I personally do think is stupid and unnecessary). But maybe the technology advances will lead to something that is useful.

      So Amazon tries this.. or that. Google does the same. They have metric ass-tons of money, and can afford to experiment.

  • "Will all this work be worth it?"

    Of course it will. Doesn't matter if every product gets stolen out of Go stores tomorrow. The amount of data Amazon is gathering about the humans who step foot into these privacy nightmares will be worth an obscene amount of money. They can sell every damn thing about you, and they will. Precise detail as to how you walk, talk, and interact with other humans around you. Your facial expressions when you pick up products. Forget purchase history. Think of every other piece of data being gathered about

  • It's costly today. What about tomorrow? (Score:3)

    by Somervillain ( 4719341 ) on Thursday July 18, 2019 @04:54PM (#58948000)
    50 years ago, a computer with the power of a pocket calculator cost an astronomical cost. A CPU with the power of the cheapest android phone was several thousand dollars 20 years ago. This is a research project. It is expensive today, but the trajectory for all these costs will go down. Computing power will get cheaper. Cameras will get cheaper. Algorithms will improve.

    On one hand, fuck Amazon. I hate that place. On the other hand, thank you for taking these risks and doing this research. I am sure it will be handy in useful (operating in underserved areas) as well as insidious ways (firing cashiers).

    The car was an insanely expensive solution that was inferior to a good horse when it started. This will improve and may take us to the next level of in-person commerce (for better or worse).

    • The car was an insanely expensive solution that was inferior to a good horse when it started

      The horse needed a barn + feed + water + groom and stable hand. The horse would be dumping 50 pounds of shit each day. Something of a problem if you were living in town,You would need a buggy or family sized wagon. Harness and tack. There would be the veterinarian's fees and any number of other expenses. The Oldsmobile 25 miles an hour. The horse 25 miles a day.

      • Eventually, yes, the car was better, which was sommervillain's point. But at first, there wasn't a convenient petrol station anywhere near. The mileage sucked, roads didn't run everywhere (so there were many useful, even near-town places that you'd want a horse), and you certainly couldn't go between cities, so you'd need to own both a car AND a horse. In other words, when the car was first coming into being, it was a poor solution for transportation and more of a rich person's novelty toy.

    • If it were just computing power theyre using it would be cheap someday but its not. The hundreds of cameras and very precise scales will not get much cheaper by default. Perhaps they are only there to train and check on the computer vision though. Regardless all of this to remove a clerk doesnâ(TM)t make sense. Even if you scale it out to remove all clerks at all grocery stores how much does that gain you over self checkout with one supervisor? They will still have a supervisor.

  • Leave with penny candy.

  • Are you a "people person" or not? (Score:4, Interesting)

    by King_TJ ( 85913 ) on Thursday July 18, 2019 @05:32PM (#58948160) Journal

    I have a theory that these automated shops will become really popular with the portion of the population who aren't that fond of random interactions with strangers.

    I think technology has done a lot to enable the shy and the introverted to communicate, share wants needs and interests, and to get business done without it always requiring "stepping outside their comfort zones".

    I say this because I've always been one of those people. Back when I first started running my own computer BBS in the late 80's, that was one of the reasons it fascinated me so much. I mean, sure - I liked computers and working with the tech. But I was captivated by the realization this was a whole new way to communicate with other people, and one I found comfortable. I didn't have to do the "face to face" thing, to get to know a person - and the communication wasn't even "real time" unless you wanted it to be (by answering a chat request).

    I've never been antisocial. I feel a need to interact with other people. But I'm just very selective about it, and I find it a very uncomfortable process going through society's expected motions with strangers (the eye contact, the small talk and so on) just to try to find that 1 person in maybe 100 worth expending more energy and time on to communicate further.

    So yes, I might use one of these Amazon stores? As I've gotten older, I've gotten better at doing all the small talk and even getting that small twinge of satisfaction that comes from having a pleasant interaction with somebody when they're a decent human being. So going to traditional stores isn't a problem for me. But I know there are people more averse to it than I am, and some who just have problems with crowds and bumping into people. (So if these stores tend to be pretty empty, that might be a selling point in and of itself, as long as that stays the case!)

    • Not sure how side-stepping a hurdle/problem is better than facing the hurdle and winning it; if someone is shy/introvert/low-selfesteem etc - whatever that bothers them to face another human/thing, it's better they invest the time/energy to gain the courage needed to win over their fear.
      • Or perhaps extroverts need to face and overcome their uncomfortableness about impersonal automated or remote interaction?

    • You must be an INTJ heh

    • Maybe I'm misreading this, but doesn't this seem kind of like you're being a bit of an asshole? Only 1 in 100 people are "worthy" of your face-to-face attention should you deem it? I'm not saying you gotta be friends with every Tom, Dick and Harry - but to suggest that only 1% of the population is worthy of you even making eye contact and saying hello or asking about the weather?

  • unexpected item in bagging area

  • is staggering. As a technological tour de force it is magnificent. In the high trust USA of the 1940's and 50's it might have been useful. In the fracturing diversity/failing trust society of 21'st C. America it has no utility for most of the country.

    By the end of the first day in (Fill in name of any city you dislike) I would expect people to swirl a scarf or sleeve or piece of paper over their hand as they reach into the display, so as to block any possible camera angle, mess up the display a little s

  • Is there anything it can't do?

  • It may work better than the article assumes since 1) employees are a cost and 2) employee theft is a major issue with maintaining inventory. The cameras and scales take care of that.

    Also, the reason Amazon won the catalog race was because they didn't need to make a profit. That allows them to be essentially a "loss leader" in any category they wish, including convenience stores.

    • Re: (Score:2)

      by DogDude ( 805747 )
      employee theft is a major issue with maintaining inventory

      If you treat them like serfs, then yes. If you treat them like humans, then no.
  • is the amazing claim Amazon makes about their Go stores.

  • yay! thanks amazon for inventing another reason to mug people and steal their phones.

  • Unimpressed (Score:3)

    by edi_guy ( 2225738 ) on Friday July 19, 2019 @12:11AM (#58949594)

    One opened up near my office and it's basically useless. Mostly prepared lunch/breakfast crap food like sandwiches, expensive bowls of fruit/yogurt. Nothing you couldn't find a block away at a Walgreens or Starbucks or whatever. I thought it would be more like a mini Target sans-registers, but the selection of non-foodstuffs is tiny. I'm pretty ambivalent about the people-quality of retail, but an Amazon Go does fee completely soul-less somehow. I won't go back, not worth it, and definitely not cheaper than the alternatives so what's the point?

    People make a big deal about the shoplifting aspect, but I think just being a Prime member is a selection bias that removes 99% of the shop lifting. Safeway wouldn't be such a dump if you had a glass gate that only opened if you had good credit...I know crappy to say, but it's true. And mind you this is not good in a societal aspect. All that sci-fi stuff is coming true slowly. Flying Ubers for the well off, the down trodden ride the stinky city bus. Whole foods for your organic quiche, Seven Eleven for your Slim Jim.

  • The world of retail doesn't really do a lot of basic research. It's good to see that someone (ie: amazon) is trying some new things in the retail sector.

    How bad is retail analysis? It's so bad that the overwhelming majority of retail "experts" said the Apple Store would fail. It's been the store with the highest sales per square foot of any store since the first store opened.

  • In my mind it would feel like shoplifting without the psychological transition to ownership that occurs at the checkout. It would feel uncomfortable for an honest person, the vast majority of people.
  • So, in the world of e-commerce, how can one run their Amazon store without opting the professional Amazon listing services [urtasker.com] ?

Slashdot Top Deals

"Everything should be made as simple as possible, but not simpler." -- Albert Einstein

Close