Episode 730: Self Checkout : Planet Money A doctor treating psychiatric patients in an emergency room created the first self-checkout machine in his spare time. Now he can't stand self checkout. So we take him shopping.
NPR logo

Episode 730: Self Checkout

  • Download
  • <iframe src="https://www.npr.org/player/embed/498571623/498627870" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript
Episode 730: Self Checkout

Episode 730: Self Checkout

  • Download
  • <iframe src="https://www.npr.org/player/embed/498571623/498627870" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript

NICK FOUNTAIN, HOST:

Goldstein, the other day you and I went to Wal-Mart.

AUTOMATED VOICE: Welcome to Wal-Mart.

JACOB GOLDSTEIN, HOST:

We got peppers, apples, bread. I got a Halloween card.

FOUNTAIN: We were there because we're doing a story about self-checkout.

GOLDSTEIN: Yeah. We wanted to find out, could we make it through self-checkout one time - just one time - without getting some kind of error message from the machine?

FOUNTAIN: We were there with Howard Schneider. He lives a couple blocks away from this particular Wal-Mart. And we were shopping there with him because he's kind of an expert on self-checkout.

AUTOMATED VOICE: Touch the item to purchase.

HOWARD SCHNEIDER: It was Fuji, right?

AUTOMATED VOICE: Move your Fuji apples to the bag.

FOUNTAIN: Everything was going fine until we hit the red bell peppers.

SCHNEIDER: Maybe it's under red? I don't know.

FOUNTAIN: So you're scrolling through the pages looking for the right thing.

SCHNEIDER: No red peppers?

GOLDSTEIN: B for bell pepper?

SCHNEIDER: Nope. No peppers.

FOUNTAIN: It was not B for bell peppers.

SCHNEIDER: Doesn't even say what to do, so then you're stuck here right now. Doesn't even say get help or anything.

GOLDSTEIN: And Fountain would probably still be standing there trying to figure this out, but for the fact that Howard went and got a store employee who walked over and saved us by typing in the code for peppers. She just knew the number in her head.

SCHNEIDER: The fact that the - we have to call the tenant to process this little order is a failure. Really, it shouldn't happen. And the machine should deal with situations like that.

FOUNTAIN: And why is self-checkout a big deal to you?

SCHNEIDER: Because I invented it, right? Why wouldn't it be?

FOUNTAIN: Hello, and welcome to PLANET MONEY. I'm Nick Fountain.

GOLDSTEIN: And I'm Jacob Goldstein. Today on the show, how an emergency room doctor created the first self-checkout machine in his spare time.

FOUNTAIN: Also, why does self-checkout still suck?

GOLDSTEIN: Before we went shopping at Wal-Mart with Howard Schneider, we went to his office to meet him. He's a doctor. He practices psychiatry. And his office is in this old shopping mall on the edge of Toronto. When he was telling us how to get there, he said, OK, my office is upstairs from the Pickle Barrel restaurant and right near this room where people come to play cards, this room called Hazels Bridge Club.

FOUNTAIN: I think this is us.

(SOUNDBITE OF KNOCKING)

FOUNTAIN: Should I try the door?

GOLDSTEIN: Hello? There is a computer - says please check in here. There's no person, it's like self-check-in.

FOUNTAIN: There's the standard beige counter where the receptionist would normally sit. But there's no receptionist. There's no papers. There's just this tablet computer facing out.

GOLDSTEIN: Yeah. So we go up to that computer and start tapping through the check-in.

FOUNTAIN: Please answer questions before you see Dr. Schneider.

GOLDSTEIN: Do you have an appointment with this person? Yes.

FOUNTAIN: It's a lie.

GOLDSTEIN: I have told the person - blah, blah, blah - has agreed to see you and/or interact with you. Please wait a few seconds.

FOUNTAIN: (Laughter) OK.

And then Howard walks in.

GOLDSTEIN: Oh, hello.

SCHNEIDER: Pleasure to meet you. Pleasure to meet you.

FOUNTAIN: Hey.

GOLDSTEIN: Hi. Jacob Goldstein, nice to meet you.

SCHNEIDER: I recognize your voice.

FOUNTAIN: Howard's in his late 50s. He's got glasses, curly hair, black pants, black polo, blue sports coat.

GOLDSTEIN: And there's something about Howard where, like, you can tell there's this gap between what's going on on the outside and what's going on in his head. There's a lot going on in his head.

FOUNTAIN: Yeah.

GOLDSTEIN: So we started out with those initial two questions. How did a doctor invent self-checkout in his spare time? And, Nick, your question, why does self-checkout still suck? And now I've got another question, which is why is the guy who invented self-checkout - this thing used in grocery stores all around the world - why is he working in a little office across from a bridge club in an old mall in Toronto?

FOUNTAIN: We crowded into this room behind the empty receptionist's area and Howard told us.

GOLDSTEIN: He said he always liked playing with computers, tinkering with stuff. He went to MIT, studied engineering then came back to Canada for med school. And he wound up working in an emergency room seeing psychiatry patients.

SCHNEIDER: I'd see on average about 8, 10 ambulances or sometimes walk-ins in the ER per day. And, you know, and you start really - thinking, I mean, you know, what's - really - what's going on in the brain?

GOLDSTEIN: As he's seeing all these psychiatry patients, he starts thinking about cognition, about what it means to think. At the same time, he's still interested in computers. So he gets this side project going. He starts dreaming up a computer that can think. And this isn't just some, like, abstract daydream. He actually writes out in, like, great detail his plan for how this is going to work.

SCHNEIDER: Silicon-based circuitry, interconnected by 5 billion vertical interconnecting wires.

GOLDSTEIN: Did you say 5 billion wires?

SCHNEIDER: Five billion vertical wires, but there's trillions of other wires.

FOUNTAIN: He submits this paper to this big NASA conference. And his paper gets accepted. He flies down to D.C., talks to a bunch of really smart scientists.

FOUNTAIN: But there's one problem with the idea.

GOLDSTEIN: So is this a thing anyone could actually build?

SCHNEIDER: It hasn't been built yet.

GOLDSTEIN: I mean, essentially even though he's written it all down, this is still some wild idea of some computer with trillions of wires and, you know, it's just a dream. It's frustrating. I mean, Howard is a tinkerer, you know, he likes building stuff. He wants to come up with something that he can actually build in the real world.

FOUNTAIN: Around this time it's the late '80s. ATMs are starting to be widespread. And IBM and these other companies decide they're going to solve the next big problem - grocery store checkout. And to Howard, it's like this weird challenge that he takes kind of personally. He goes and tells his wife - this is my big project.

SCHNEIDER: And my wife says you've never even gone to the grocery. You refuse to go shopping. You get very upset going to shopping malls.

GOLDSTEIN: Howard, our hero, is not deterred. He walks over to his neighborhood supermarket to get some data.

SCHNEIDER: And I had a clipboard. I had a stopwatch - mechanical one.

GOLDSTEIN: You got the mechanical stopwatch, like, tick, tick, tick, tick, tick?

SCHNEIDER: Yes, yes, yes.

GOLDSTEIN: Standing there at the supermarket, he sees a supermarket cashier isn't just some simple robot like an ATM. Being a cashier is more complicated than it seems.

SCHNEIDER: Supermarkets are messy environments. You have things without barcodes. You have produce. You have people trying to steal things. People are saying give me this, make change, do this. And that's a huge leap in cognition to go from banking machine to replacing the person in the supermarket.

FOUNTAIN: Howard thought, OK, this is difficult, yes. But it's also exciting. Here's a thing I can actually build. And it combines all the cognition stuff that I've been thinking about at the hospital and all the computer stuff that I've been working on as well. So he sets to work, takes all the money that him and his wife have saved up - it's about $50,000 at the time - and starts to build a prototype.

GOLDSTEIN: You know, and so he's building his machine. And some pieces are pretty straightforward, you know, reading barcodes - that's just a scanner. Taking the payment, there's a machine that does that. I mean, to do those things, a computer doesn't need to be aware, you know, it just needs to do what a computer does. But there is a place where the computer does need to be aware, does need to be thoughtful.

FOUNTAIN: Yeah. It's theft, right?

GOLDSTEIN: It's people stealing stuff, right? Hey, there's no cashier (laughter). I'm just going to walk out with half the stuff in my cart. This is a very hard technical design problem. You know, it's like if you're building this machine, you've got these sort of dials you can turn - right? - like you can make it really easy to steal stuff. And then people are just going to steal stuff. Or you can make it really hard to steal stuff, but then you're going to end up accusing lots of honest people of trying to steal stuff.

FOUNTAIN: Yeah. This is why I hate self-checkout, right? It's that line - unexpected item in the bagging area. That phrase is burned into my memory. And then there's that moment, right? You're shopping and there's a big lot of people behind you and you're trying to find red bell pepper on that little screen. But you just want to be done, so you pick green pepper. You might save a buck. And is it stealing? I don't know - maybe.

GOLDSTEIN: Yes. Yes, it's stealing.

FOUNTAIN: It's stealing.

GOLDSTEIN: So Howard Schneider is confronting all of these problems as he's trying to build his machine. And people around this time have already figured out that you can put a scale in the bagging area to try to prevent theft.

FOUNTAIN: Yeah. And the reason behind that is simple, right? They want to make sure that people don't put stuff in their bags and walk away with them that they haven't scanned.

GOLDSTEIN: It's a simple idea, but Howard found out it's a really hard technical problem to solve for a few reasons. Partly it's because stuff at the grocery store comes in this huge range of weights.

SCHNEIDER: They sell greeting cards and they sell 40 pounds of flour or potatoes.

GOLDSTEIN: So Howard has a scale custom-built that can handle everything.

FOUNTAIN: But even then, he has another problem. And that's that the weight of the same kind of item can vary. You know the loaves of bread they bake at the store? They're a little bit bigger or heavier on some days, and a little bit lighter on other days.

GOLDSTEIN: So Howard programs his machine to figure this out, to actually learn as it goes along.

SCHNEIDER: After the machine's seen the loaf of bread a few times it realizes, wow, bread can vary. On the other hand, Coca-Cola, they're pretty good. The Coke cans are within, you know, within a gram or two each time.

GOLDSTEIN: It's not perfect obviously, but it's good enough. He puts it all together - the good enough scale, the scanner, the monitor - and he's got his machine. He's ready for it to take over the world. Howard showed us a picture of this first machine and it's really striking. It looks almost exactly like self-checkout machines still look today.

FOUNTAIN: Howard still doesn't quit his day job. He's still working at the ER. But he does turn his self-checkout out hobby into an actual company. He calls it Optimal Robotics - very '90s - and he starts going around pitching to grocery stores.

GOLDSTEIN: He gets a meeting at a New England supermarket chain called Stop and Shop. So he drives down. He gives this presentation in this boardroom full of people. And at the head of the table there's this executive who everybody's looking to.

SCHNEIDER: This old guy sitting at the head of a table and, you know, he says, son, do you know why people shop at Stop and Shop? I said no. He says because of our people, because of our people. And then everybody around the table starts nodding their head. And he says, do you think people want to come in and speak to your machine?

SCHNEIDER: Do you know anything about retail? Do you know anything? And he says, I don't think it's for us. And then everybody around the table nodded their head. And I drove back to Montreal with my tail between my legs. And it was just a horrible feeling thinking this is the stupidest thing I've ever done. Nobody will ever like it or ever use it.

GOLDSTEIN: But Howard keeps loading his machine into the back of this rented Ryder truck, keeps going to supermarket trade shows. And eventually he gets a meeting with a supermarket chain called Price Chopper. It's based in upstate New York.

SCHNEIDER: The head person actually saw the machines. And, you know, he said - he looked at it, he says, I like it, let's do it. And guess what? After that, everybody's very nice.

FOUNTAIN: The CEO says, we're not going to buy the machines from you, but you can use one of our stores in upstate New York as a real-world experiment.

GOLDSTEIN: And on August 5, 1992, grocery store shoppers use what may be - depending on your definition - I mean, we - what we are going to call the first fully-automatic self-checkout machines.

FOUNTAIN: When Howard gets to the supermarket that day, he sees that there are protesters there. It's the cashiers union. They're saying these machines are taking our jobs. Howard read us this quotation from a union rep at the time. And the union rep basically said this thing that I think about all the time - these machines are forcing customers to do more work.

SCHNEIDER: Quote, "maybe if the customers come early, they can unload the truck and stock the shelves," unquote.

GOLDSTEIN: Well, I mean, like, there is a serious question underneath it, right? Which is, like, are these machines taking people's jobs?

SCHNEIDER: But the answer's very anti-climactic. Let me tell you what happened to Price Chopper. The machines went in, people used them. And they didn't work - it crashed. It stopped working. And then people had shopping carts half unloaded and everything stopped.

FOUNTAIN: The store is full of angry shoppers. And Howard, he's just sitting there in the middle of all this chaos, unscrewing the back of the machine so that he can restart it. And the union, they figure they've got nothing to worry about. They pack up their protest and leave. But the store manager - his name is Rick (ph) - he's not happy.

SCHNEIDER: He says, I don't know, if these machines crash, I'm throwing it out of my store - literally. I said, it will work. I have no idea why this happened. This shouldn't have happened. The machines are perfect.

GOLDSTEIN: (Laughter).

SCHNEIDER: I mean...

GOLDSTEIN: Those two statements - it will work and I have no idea why this happened - seem sort of at odds with each other.

SCHNEIDER: Yes, yes. I would agree. I would agree.

GOLDSTEIN: Howard stays up all night sitting there in the grocery store fixing the code. He says he can still remember smelling the cinnamon buns, they are baking in the back of the store while he's sitting there trying to figure out what's wrong. And the next day, his machines didn't crash. But the shoppers and the machines, they still needed a lot of hand-holding. In fact, they need so much help that Howard has to invent something else. He has to invent the person who stands there next to the self-checkout machines, helping the customers and fixing the machines.

SCHNEIDER: We drilled some holes in the cabinet so that if you had to reset the machine, you could just stick the long screwdriver in...

(LAUGHTER)

SCHNEIDER: ...And hit reset button. So we gave him a long screwdriver. And he was very friendly, very nice with people. And he would, like, he'd help them through the orders. And that made a huge difference.

FOUNTAIN: It's clunky, but it's working. The machines are getting better. And then Howard gets what looks like it's going to be his big break. Kroger, one of the biggest supermarket chains in the country, wants to start buying his machines. And he starts dreaming of this very big future. We asked him back then, when you imagined the grocery store of the future, when you imagined the grocery store of, say, 2016, what did it look like?

GOLDSTEIN: So in that grocery store of your imagination, are there any human beings stocking the shelves?

SCHNEIDER: No.

GOLDSTEIN: Are there any human beings doing checkout, being cashiers?

SCHNEIDER: No.

GOLDSTEIN: Are there any human beings working there at all?

SCHNEIDER: Yeah. You have some people taking care of the machines and dealing - somebody wants to see a human, so a human can come see them.

GOLDSTEIN: Maybe one token human.

SCHNEIDER: Well, kind of more than that. We don't like to use the word token. We just, you know, yeah.

(LAUGHTER)

GOLDSTEIN: So here we are. It's 2016, and the grocery store of the future looks a lot like the grocery store of the past. It looks basically like grocery stores looked in 1992. About 90 percent of checkout lanes in U.S. grocery stores are still manned by humans. What went wrong?

FOUNTAIN: Yeah, we talked to a few people about this. And a lot of the things that Howard was dealing with in the early '90s are still big problems today. Just last week, there was this story about a police crackdown in Australia on shoppers stealing stuff at self-checkout. There is apparently this one store in Sweden that's entirely automated. But, like, of course, yes, we know, Sweden, you are perfect, you've never stolen anything. Not going to happen in the rest of the world.

GOLDSTEIN: In the rest of the world, there is a place where self-checkout is great. In fact, it's so good that we don't even think about it as self-checkout - online. You know, every time you buy something from, say, Amazon, you go through self-checkout. There is no cashier and it's super easy.

FOUNTAIN: It's also really hard to steal from Amazon.

GOLDSTEIN: True. I mean, you could say, sure, sure self-checkout out from Amazon is easy, you know, they don't have to deal with thieves. They're not asking you to look up a bell pepper that doesn't seem to exist in their system.

GOLDSTEIN: But that is my point - it works. It doesn't have all of these problems. You know, when you think about grocery store self-checkout, I think there's a good chance that it will continue to suck for as long as it feels like self-checkout, for as long as it feels like you are doing this work, this thing that a person used to do. I mean, pretty clearly I think in the long run, the real solution is not a better self-checkout machine - it's no checkout at all.

FOUNTAIN: We ran this idea by Dusty Lutz. He runs the self-checkout group at NCR, that's the corporation formerly known as National Cash Register.

GOLDSTEIN: Dusty Lutz is kind of a big deal in the self-checkout world. In fact, he was in the room when they came up with what is probably the most famous phrase in self-checkout.

AUTOMATED VOICE: Unexpected item in bagging area. Remove this item...

DUSTY LUTZ: I remember very clearly the day we actually came up with that because we were trying to figure out, how do we communicate that you might be stealing something but you might not be stealing something? So how do we communicate that? And there were many, many things like stop, thief, remove this item. Don't steal this from us.

(LAUGHTER)

LUTZ: But in the end...

GOLDSTEIN: What do you think you're doing? You're not fooling anybody. We're watching you.

LUTZ: Yes. Yes.

GOLDSTEIN: He said there are already some stores where you can pick up a scanner when you walk in and scan your stuff as you walk through the store. But that, he said, is just a transitional phase. Here's what the store of the future looks like.

LUTZ: You walk in the store. You pick up your items. You walk out.

GOLDSTEIN: No cash register at all, no cashier?

LUTZ: That's right.

FOUNTAIN: You've already registered your phone with the store, so they know who you are. And every time you pull something off the shelf, there's some tag on the item or some camera that recognizes exactly what you're putting in your cart. And then when you walk out the door, the store charges everything that's in your cart to your credit card.

GOLDSTEIN: And do you think that'll actually happen?

LUTZ: That will happen, yeah. The question is will that happen in 2020 or will that happen in 2030?

GOLDSTEIN: So that is the past, present and future of self-checkout. There is one last question. Why is the guy who invented all this, Howard Schneider, still working in a little office across from a bridge club in a mall in Toronto?

FOUNTAIN: Howard is really good at thinking about the future. And he's really good at tinkering with machines. But he's not really good with thinking about money. Howard told us back in the mid-'90s, just as he was getting this big order from Kroger that was going to be his big break, he had this realization.

SCHNEIDER: I said, wow, you know, if I start buying more parts, I'm bankrupt. I couldn't afford anything. You know, my wife's saying, oh, you're spending every cent. Because every cent I made in the emergency room, I spent on that. And, you know, we had a 1-year-old-son, like, she has no money for milk. This is, you know, what kind of father are you? What's going on? And I realized, you know, what have I done? I solved all these problems, but I guess I didn't really do the financial problem well enough.

FOUNTAIN: We should say we asked Howard if we could talk with his wife. He said she didn't want to talk with us.

GOLDSTEIN: Just as Howard was about to run out of money, an old friend of his stepped in and basically saved the company - raised a bunch of money.

FOUNTAIN: Howard told us he was embarrassed by his financial failure. He told us he felt like it wasn't really his company anymore and he left. In 1995, he sold his stake in the company, says he got enough money to buy a nice house and a nice car in Toronto. And that was it for him and the company.

GOLDSTEIN: In 2004, the company that Howard had founded sold for $30 million.

FOUNTAIN: Howard says he regrets leaving when he did, and not really because of the money.

GOLDSTEIN: Yeah. He really doesn't seem to care about the money. The reason is he thinks if he stayed, he could have built a lot more stuff.

SCHNEIDER: You'd go to the airport, they'd be all my machines. You'd go to a restaurant, my machines would be everywhere.

GOLDSTEIN: Lately, Howard has been working on a new project. He calls it the DocPod. It's a chair. You sit in it and it takes your vital signs and sends them to a computer. He built this prototype in this, like, egg-shaped chair. It's white plastic on the outside and red velvet on the inside. It looks like a chair from "The Jetsons." While we were visiting Howard, he sat in it. He looked happy.

(SOUNDBITE OF MUSIC)

GOLDSTEIN: We do love a good origin story here. If you've got one, some gadget in everyday life that has an interesting or surprising story behind it, tell us about it. Maybe we'll do a show. You can email us at planetmoney@npr.org. Or you can find us on Facebook or Twitter - we're @planetmoney and Nick Fountain and Jacob Goldstein.

If you're looking for something else to listen, to listen to the Fresh Air podcast. It's great. It's Fresh Air, but you can listen to it anytime you want. Recently, Terry Gross interviewed David Fahrenthold of The Washington Post. He's been doing really interesting work looking into the Trump Foundation's finances. You can find the Fresh Air podcast on the NPR One app at npr.org/podcast or wherever you get your podcasts.

FOUNTAIN: Thanks to Neil Weschler (ph). And thanks to our producer, Elizabeth Kulas. And also, thanks to our intern - Reyna Cohen (ph). We're actually accepting intern applications right now. We're looking for one for the winter and spring semester. I once was an NPR intern. You should be one, too. find out more about it on our website. I'm Nick Fountain.

GOLDSTEIN: And I'm Jacob Goldstein. Thanks for listening.

Copyright © 2016 NPR. All rights reserved. Visit our website terms of use and permissions pages at www.npr.org for further information.

NPR transcripts are created on a rush deadline by Verb8tm, Inc., an NPR contractor, and produced using a proprietary transcription process developed with NPR. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.