Loading summary
Transcript81 lines
- [00:00]
A
Foreign.
- [00:10]
B
Hello, welcome to Slate Money Food, our special Slate Money mini season on the economics of the food industry. This week we are talking about technology. Yes, technology has invaded even the food industry, which is on some level depressing. We're going to talk about just how depressing that is and just how sophisticated and complex the tech stack is in today's restaurant industry. With Jordan Thaler. Hi, Jordan.
- [00:40]
A
Hey, Felix.
- [00:41]
B
Tell us a little bit about who you are and what you do.
- [00:44]
A
So I started a company in 2011 that ingests point of sale data from tens of thousands of restaurants. And we use data science to prescriptively fix problems in the restaurant business.
- [00:56]
B
And what's the name of the company?
- [00:58]
A
What's Busy?
- [00:59]
B
So, Jordan Taylor from what's Busy? Coming up on Slate Money Food. So, Jordan, you are an expert on delivery, which is this new big thing which is only getting bigger and more timely.
- [01:14]
A
Right now, expert is maybe not the right term, but we see a lot of data. We definitely see a lot of data working with a lot of POS companies.
- [01:23]
B
Okay, so first, what is a POS company?
- [01:25]
A
The POS is an acronym for point of sale. I know you were thinking something else. And it's the, the nerve center of a merchant's operation. So traditionally it's where all the sales data stored. It's where employees clock in and clock out. It's got all the information you send to accounting and payroll, and it really is the central nervous system of the.
- [01:44]
B
Business and you manage to get access to all of that information.
- [01:49]
A
Mostly what you find is the POS industry in general is incredibly diffic. So, I mean, it's a herculean task, really. It's like Sisyphusian. You are constantly doing inordinate amounts of work to build this brain of a local merchant and you get paid peanuts for the effort. And over time, many of the POS companies realized they had access to all this really rich, useful data that merchants and in general business ecosystems never had access to. And they would find us and say, hey Jordan, we have this data. We know in theory it's valuable. We just don't have the resources or know how to put it to work and execute something meaningful. So you take this and please do something for us. And that's how we've persisted over the past eight years.
- [02:33]
B
Okay, so you, you have, you know what's going on. You've seen what's going on. So let's talk a bit about trends. Is it true we all believe that everyone is grubhubbing and seamlessing And Uber Eatsing and basically getting food delivered way more than they ever used to in the past. Is this true?
- [02:51]
A
No, it's true. I don't know how much that is really economic. I think with the Fed rate increases or early decreases in 08, you saw the shore up of private capital in the order of 2 trillion or something. Last numbers I saw. And that pushed a lot of companies to raise, in my perspective, obscene amounts of money to chase the idea of delivery. And you have a lot of subsidies. And so consumers aren't paying the true economic cost for delivery. And I think as you have market pressures come back into reality, you might see those economics become a little more transparent and the consumer might not enjoy paying the true cost of that service.
- [03:27]
B
So, so, so basically what you're saying is that Uber's been losing billions of dollars, grubhub's been losing billions of dollars, and all of those billions have essentially been going to help subsidize cheap delivery for people getting delivery. And if those billions of dollars of subsidies go away, then maybe the spike in delivery might go away as well.
- [03:47]
A
Well, GrubHub, I believe, has been profitable every quarter since 2015.
- [03:52]
B
So how is GrubHub making money? If Uber Eats is losing money, what's the difference between them?
- [03:57]
A
My understanding is that grubhub primarily was an online ordering business. You don't have to get into the mess of last mile logistics and delivery, and they would still charge a pretty significant fee as that service provider. Uber and DoorDash really forced their hand into delivery, which is not a very profitable business. You can go look at FedEx as a comp, but GrubHub got into that business. I don't know if they break it out by delivery versus online ordering, but my suspicion is that the delivery business on a unit basis is actually not profitable at all.
- [04:24]
B
But for the time being, so long as the lovely venture capitalists are willing to throw money at us, we are ordering in much more. That's true.
- [04:34]
A
Yes, 100% true.
- [04:35]
B
So, yeah, give us, give me an idea of how much more common is it now than it was say five years ago for people to get restaurant meals delivered.
- [04:46]
A
So what you find is roughly 30% of a restaurant's business has always been we call off premises. It was just typically someone picking up the phone and calling in an order. Now that segment of the population saying, well, I could pick up the phone, but I'm lazy. I'll just get online and click on some buttons and do the same thing, the same net effect to me, except I do it digitally. So I think. I wouldn't say that the off premises is growing so much as it's cannibalizing the phone orders. So it's still a material part of a restaurant's business. That 30% really seems unchanged. It's just that more and more of that seems to be emanating from the online providers.
- [05:22]
B
So that. That's surprising. And the phone orders, they would generally be delivery as well. They wouldn't be like pickup.
- [05:28]
A
Yeah, it'd be. It'd be pickup. It would be pickup. Someone calling in. Yeah, someone calling in and saying, hey, we've got. I Want to order 5 items from your menu.
- [05:36]
B
And then I would drive down there in my car and pick them up myself.
- [05:40]
A
That's right.
- [05:40]
B
And trying to pay someone to drive them to me instead of me doing the job of driving down there myself obviously significantly increases the cost of the restaurant of providing that service. Because if I do it, it's free.
- [05:53]
A
Yeah, that's right.
- [05:54]
B
So we've done something which was weirdly sort of economically sensible, which is that people would take that driving bit of the equation onto themselves because they didn't really feel the cost of it. And we've turned it into a significant cost centre which is draining billions of dollars out of venture capitalists pockets.
- [06:14]
A
Yeah. I think the music stopping. I think you're seeing doordashes being pushed to go public. You're hearing about backroom talks with mergers and acquisitions between these guys that are all losing money. I think the private markets have said, hey, we've given you enough capital, you need to go figure out the rest by yourself. My suspicion here is that consumers aren't going to bear the cost of the true economics of delivery. And so it's going to be a service that's relegated to much smaller parts of the market for people that can afford to pay $10 to have a $5 hamburger delivered.
- [06:48]
B
So tell me about cloud kitchens and this idea that it is possible to make good money by setting up a sort of restaurant in the back of a warehouse somewhere and just delivering everything and having. And having 100% off premises.
- [07:02]
A
It's great in theory. I think scale is going to be the true test of whether this works. Part of the what I suspect. Right. I'm not in these meetings, but you've seen leaked information. Like I know the leaked deck from Deliveroo definitely hinted at this, which was, look, we're losing money as delivery providers. I'm uber eats, I'm losing tons of money, but if I verticalize the restaurant production. Meaning I open this ghost kitchen, I look at my data and I say, hey, I see that chicken pad Thai is selling really well in this zip code for this price point. Let me just go ahead and make it in my kitchen myself, call an Uber Eats pad Thai and put it on here for 10% less than the nearest competition. Can I sort of suck up that market? Right. Become if my belief is that food is fungible, it's fairly commoditized, and can I reach 8020 in quality and be the lowest cost producer through scale and volume and just win this category? That's a thesis, right? We don't have enough data to really know. We've seen the stuff that Kitchen United has put out. We've seen some other data points and it definitely appears at a high level to make sense. But that's like the Mike Tyson, right? Everything makes sense to get punched in the mouth. So who knows what's really going to happen?
- [08:17]
B
So let me take that apart a little bit and let's start with this idea that food is fungible and consumers order on price. Is that something which has been. Which you've seen in the data?
- [08:30]
A
We've not looked at that deeply. Right. I definitely think you find cohorts of consumers that are price conscious and they think about do I really want this or my will I find a substitute that seems to be just as good in quality for a lower price, but we've really not gone deep enough to the consumer part of the equation to understand what's really happening there.
- [08:50]
B
Presumably if consumers were that price sensitive, then they would still just be ordering for pickup rather than paying like 10, 15, $20 for all of the various service and delivery fees that get slapped onto your typical grubhub order. Right. I mean, it seems that one of the weird things that we've seen in terms of this explosion of restaurant delivery services, that it can sometimes weirdly be more expensive to get a meal delivered than it is to go out and sit down and maybe even order a glass of wine and have a decent meal delivered to you by a server.
- [09:26]
A
Yeah, that's definitely true. I think that's right. You're also framing things from a very logic oriented worldview. I don't know. Most consumers aren't very good at math, so I think it takes them a while to figure stuff out. But you're right. If you do the math and you sit down and you put pen to paper, it's like, wait a minute, maybe I should just go into the restaurant and get my meal and pay a tip and get a glass of wine because it will be cheaper.
- [09:46]
B
So that, I mean, I feel like that's good news for those of us who like the idea of restaurants, who like the idea of ordering from small businesses and local maybe like, you know, immigrants into the community who have great food and recipes and we want to support them, that somehow they're not in such a bad position as you might believe. In a world where the intuition is often, well, like the big online giants are just going to steamroller everyone with their marketing budgets and their delivery and their logistics and that kind of stuff.
- [10:21]
A
I mean, it's definitely challenging food service, you're manufacturing perishable goods. It's a very non trivial operation, which is why you have such massive fragmentation. Even McDonald's, which may be the largest operator by unit, counter by volume, they're still not very big. From a total market perspective, 15,000 U.S. units out of, let's call it 650,000. So they're not massive. It's not Walmart, who's got 25% of the grocery share. I think that there's a certain segment of the restaurant population that is, I would say nearly insulated from the ghost kitchen and the delivery model. There are white tablecloth restaurants that are $50 per entree where you're just look, I'm not getting that delivered. I'm ruining the experience, I'm ruining the ambiance. Why? Right. So there are establishments in that category and it's probably a dwindling category that are fine. It's a lot of the casual dining brands have, from the consumer's perspective, a commod monetized offering where I think there's trouble. And you've seen the burgeoning of the fast casual category in general as a testament to the change in consumer eating patterns. And it's been, it's been stressful for the casual dining brands. They look at that and say this is a real, a real assault on our business model.
- [11:40]
B
Okay, so help translate these things for me because I'm a New Yorker who doesn't understand brands. What is the difference between a casual dining brand and a fast casual brand?
- [11:50]
A
A casual dining would be like a Ruby Tuesdays or Applebee's. And a fast casual would be the up and coming sexy concepts like a Zoe's Kitchen or a Cava. These newer brands that are a lower price point, it's typically healthier food, it's a faster delivery model. So the real difference is if you go to a casual dining restaurant, you sit down and someone serves you a fast casual is you might go to a counter and order and they'll call your number and you can get the food.
- [12:15]
B
So in terms of, if you think of a sort of spectrum, maybe you get like old fashioned fast food at one end and then first casual and then casual dining. And all of this has become what you might call like brandified, right. Is it fair to say that in most of these categories you're really talking about large companies with brands and marketing budgets in multiple locations?
- [12:38]
A
I think how you define enterprise, let's just call it like 10 or more locations, right? Those enterprise type merchants, they represent about half the market. So half of the 650,000 locations are chain or enterprise groups and they're. The rest are SMB. There's definitely economies of scale. When you get in to become a larger restaurant, you get middle management, you get purchasing power. There are things that work in your favor and they're probably branding things that work against you. Like yeah, you might have a large marketing spend, but your food quality is probably not on par with the restaurant owner who's got one shop that focuses on that all day long.
- [13:14]
B
And tell me about the relationship between the enterprises, especially the very big ones. The McDonald's of this world on the one hand and the last mile delivery people on the other. The grubhubs and the eats and those guys.
- [13:27]
A
It's been comical to be honest. So you had these delivery providers believing that what they made was so incredibly special and couldn't be replicated. And you would see them signing exclusive deals with these large enterprise restaurant brands and then after six or nine months, the restaurant brand would realize, wait a minute, there's nothing special about what you do. You guys are the definition of a commodity. It's about pricing. And so I'll start to work with a number of you providers. McDonald's has gone that way. I think they were exclusive with Uber Eats and now they work with a number of third party delivery provider. So I really do think it's a massively commoditized space because venture capital or just private capital in general has been so easy to come by. People said I'm willing to bet my money in a fund that someone else can replicate this model. And if I give them a couple hundred million, or in some cases billions, they can go ahead and be the market winner. That's what you have effectively. I think this is more an issue of just too much dry powder being in the markets than anything else.
- [14:24]
B
It does feel a little bit to me like Amazon I order something on Amazon prime and then it arrives. And sometimes it arrives via FedEx and sometimes via UPS and sometimes via USPS. And it's like, I don't care.
- [14:35]
A
Yeah.
- [14:36]
B
And I don't get any brand loyalty to the brand that's making that delivery. They're just competing on price. And Amazon will use presumably whoever is cheapest. Right?
- [14:45]
A
Yeah, that's right. And Second Measure, which analyzes a panel data from credit card spend. They have some great metrics out there. They work closely with the Journal and they show routinely that customers have no loyalty to an Uber eats or a DoorDash. And they will just go between whoever got the best promotion that week and that's who they use.
- [15:03]
B
I mean, I guess that's the big idea, right? Is that uber eats and DoorDash and Grubhub, and those guys are all like saying, we want people to use us as their way of ordering food. If you're sitting at home, you say, I want food, then you call up Seamless, you call up DoorDash and you pick a restaurant from there. And if you're going through that app, then by definition, you're going to use that delivery service. What you're not doing, it seems, is saying, I want to get McDonald's or I want to get Chipotle, or I want to get like such and such specific restaurant down the street, and then ordering directly from them, and then that restaurant gets to pick whichever one is cheapest.
- [15:42]
A
I think if a consumer has a strong brand affinity, meaning they know they want to order from Chipotle, they might order from Chipotle directly to get loyalty points or rewards points or whatever that program is. I think most consumers don't do that. They're like, I'm in the mood for XYZ cuisine type that I can arrive here within 30 minutes at this price point. And then I shop. And I was at an investor dinner in New York a few months ago where one of the investors said, look, I have all three apps on my phone, the DoorDash, the Uber Eats and the Grubhub. And I find a restaurant and I just see who's got the lowest price for that restaurant. And that's the one that I order from.
- [16:19]
B
Wow.
- [16:20]
A
Yeah. I think there's no. The consumer has zero loyalty to any of these programs.
- [16:25]
B
I'm thinking in my head about the best performing stock of the past 10 years is famously Domino's Pizza, that they've managed to sort of reach invent themselves as this technology and delivery company. Are they an exception in terms of sort of vertical integration somehow?
- [16:43]
A
Well, yeah. So for starters, yes, they can vertically integrate because pizza travels well and their model's always been heavily delivery component. Right. So they have those economics well studied and have it figured out. The other thing going for them is the restaurant entry in general is massively unsophisticated. Like the idea of data science in a restaurant goes way over the heads of most operating executives, even at large public restaurant companies. It's pretty similar in retail, too, which is why Amazon's done so well. Domino's has done a very good job of building a culture on data and engineering. And they spend money and they prioritize a lot of those efforts to eke out the nth degree of value from whatever service or product they're making. A lot of other restaurants or competition, they don't have that culture. And that's been a big problem. I think over time it changes. One of the bull bets on the ghost kitchen model. So why Travis was able to raise 400 million for his cloud kitchens concept was, hey, cloud kitchens will use data and be much more intelligent about pricing and logistics and everything else. And that will in turn yield better margins for investors. And there's nothing preventing a large restaurant chain from doing that themselves. They just don't have the culture. It can't be overstated enough. It's really hard to attract good data scientists, good engineers, if they come to work for an organization that doesn't care what they think, doesn't care what solutions they come up with. If that's not prioritized, if you're not prioritizing on being a data operating company, you're just not going to attract that talent. And that's the real reason why Domino's wins, is they've done a good job with that culture.
- [18:22]
B
The company that no one's ever heard of in this space is the. The way that I actually discovered you in the first place is olo. Who or what is olo?
- [18:30]
A
So, POS companies, if we go back to how we started this conversation, pos companies oftentimes realize how little revenue they earn and how hard it is to rip out the point of sale because it's the central nervous system. And so they will often chase tangential features or products. I call them bolt ons, to increase revenue for their core business. One of those has always been online ordering. The POS companies just have never been able to do online ordering, let alone pos, at a very respectable product output. OLO recognized the opportunity of online ordering. They saw it was going to be a massively popular and growing market. And they said, look, we can just build online ordering really, really well and be best of breed. And at some point the restaurant operators who all use a point of sale will recognize that their point of sale can't do this the same way we can. And they'll need to use our services, especially as their customers are demanding more online ordering. And that's exactly the niche that olo's carved out. They've become the best of breed play in online ordering. And they're in like 70 or 80,000 restaurants because of that.
- [19:40]
B
And they don't do the delivery. They just literally they are the layer in between my computer and the cash register, basically inputting my order into that register.
- [19:51]
A
They've done a little more than that. So they've built this product called Dispatch, which I want to say Noah announced on Jim Cramer show Mad Money two or three years ago. Which sort of is this marketplace of if you use the core OLO service for online ordering and then that customer wants a delivery, Dispatch can route that delivery to any number of third party delivery providers and sort of market make and choose the provider that's going to have the fastest turnaround time, the lowest price, et cetera. And that's been something that's been pretty popular with a number of large chain restaurants. And I think Noah's been very visionary in seeing how that's going to grow.
- [20:26]
B
Noah being Noah Glass, the CEO and founder.
- [20:29]
A
I have never met somebody in my time in this restaurant industry who has worked as hard as Noah to build something as powerful and as future thinking as olo. He's finally reaping the rewards. And I tip my hat to the guy. He's just, gosh, he's put in 14, 15 years to get it to where it is and he's a saint.
- [20:49]
B
So what else do we need to know about this space? Because it does seem to be this sort of brave new world of tech stacks and engineers and things which have to plug into other things and big data. And it seems so many miles away from what most people think about when they think about a restaurant, which is like you order some food and they make you the food. Presumably. There's a lot of different moving parts here.
- [21:15]
A
There are. I think one of the biggest components that I think merchants will need to be cognizant about over the next few years is payments processing. So payments processing is an incredibly lucrative part of the stack now for service providers. So you probably know this to some extent, but you go to a restaurant, you Swipe your credit card, you spend $100. Only 97 end up in the restaurant's bank account. The other 3% go to a number of different parties. And the problem as we've seen, is the payments ecosystem. The folks that earn that 3% do so in very ambiguous ways. Some of them border on malfeasance. Frankly, I think restaurants have been fleeced on this for a long time and the point of sale and technology models are converging on payments. So building software is hard. Doing payments is like your hamster could do it. So the software providers and the vendors are realizing, hey, we need to get into payments. And some of those providers are doing so in a what I would just call an unethical manner in how they portray the fees and how they portray the value. And merchants who aren't any sophisticated parties really don't know how this works and are being taken advantage of and it's just a non trivial component of that entire stack.
- [22:32]
B
And as the world becomes more technological online, requiring a certain level of sophistication when it comes to data science and tech stacks, does that give more and more of a natural advantage to the 50% of the market that's enterprise versus the 50% of the market that's mom and pop.
- [22:55]
A
Yeah, I think it does. So the biggest challenge for if you were to start a data science outfit today and sell some really great products and target merchants, it's just a lot easier to do distribution to larger chains than it is to the SMBs. And you have to look at your customer acquisition costs and at some point you'd realize, I just can't justify selling to the SMBs.
- [23:16]
B
And even if I could, the SMB probably wouldn't have the necessary sophistication to be able to use it. I remember Talking to the OpenTable guys many years ago. They were like talking about how incredibly powerful data from their reservations platform could be to restaurants. And it turned out that none of the restaurants who used their software managed to use even like 10% of the capabilities. Because that's not what restaurateurs think like that. That's not how they think.
- [23:44]
A
Part of the blame lies with the open tables, right? So they're not doing a good enough job with their product to make it easy enough to use to show the merchant, hey, here's why data matters and make it simple to use, right? Instead it's usually like, hey, here's a bunch of reports. I'll pretend you're an engineer too and you go figure this out. And that's not how it works. The other problem is merchants in general are very frugal and they don't pay for value. And so you could build a really good solution, but if the merchant doesn't pay for it, what's the point? So you kind of have these battling positions of yeah, you could blame the vendor, but you also got to blame the merchant. If they're not going to pay for value because they don't realize it, then what's the point? To spend the R and D capital to build a better mousetrap.
- [24:28]
B
That's where things like franchise operations have an advantage. It's not like you need to persuade every franchisee to spend money on this. You sell it once at the corporate level and then the corporate level makes it back through compulsory payments from the franchisees. And the franchisees get the value, but they don't have to make the decision.
- [24:45]
A
Sometimes it just depends on the organization. You'd be shocked at how many franchisors don't even require their franchisees to standardize on point of sale systems. And you ask, well, why does that matter? Well, if you don't have unified data coming from all your locations, how do you know if marketing is working? How do you know who's got labor problems? These are things where you need standardized data to answer these questions. And a lot of franchisors just still haven't figured this out.
- [25:08]
B
So from a tech perspective, who are the most sophisticated franchisors out there?
- [25:15]
A
You know, we don't sell direct to enterprise merchants, so I'd be a really bad person to opine on this. We just look at the data, right? So we partner with the POS companies, we see the data in general, I think that everyone is really unsophisticated. I mean, we've seen data from again, large public brands. Sometimes we're brought in by activist investors and it's a disaster. These guys have no idea what they're really doing.
- [25:35]
B
I mean, that's a little bit reassuring to me. On some level, I kind of feel like running a restaurant on some level has always been a relatively kind of low tech thing. And I kind of want to live live in a world where restaurants are low tech things and they're not. They don't require a huge amount of technological sophistication to be able to run. I want to be able to walk into a place and just say, you know, make me something delicious. And then that's their skill set is making something delicious and they serve me something delicious. But maybe I'm just being Too much of a romantic here.
- [26:05]
A
I don't think that sentiment is wrong. I think the complexity of running a business today is much higher than it was even 10 years ago because the market is so saturated. And so really the goal of technology is not to get you in, in the back office on your computer, figuring out what to do with spreadsheets. It's to automate parts of your business so you can focus on the guest experience, so you can deliver that meal and that expectation that the customer came in with when you walked through your door. And technology providers in this space generally have not fulfilled that end of the bargain. And again, part of it's their fault, but part of it's the operator who just doesn't want to pay for technology.
- [26:42]
B
I have one last question for you, which is the. The promise of the grubhubs of this world, if I am a mom and pop and grubhub comes up to me and says, listen, you have a handful of people who are phoning you up and asking for delivery or asking for takeout and you'll get those. But then if you sign up on my platform, then a whole bunch of people who just go to GrubHub or pull up the GrubHub app instead of knowing about you, we'll be able to discover you and we'll order stuff from you. And that's purely additional. Like it's just free, extra custom for you and almost so long as you kind of even break even on that. It's just more volume, you get to make more money, higher revenues. And that's just a brilliant offer which I'm offering you. Like is that true or is that like deeply misleading?
- [27:35]
A
So I think it was probably true in the early days. Now it's no longer a secret. Right? And everyone's competing for a top rank on grubhub or Uber Eats or just like they compete on Google. So it's just one new category with which to compete.
- [27:51]
B
I remember reading an amazing article once about the top rated restaurants on TripAdvisor. And there's like a few restaurants at least in New York City, but also in other cities where they have completely worked out how to get that coveted number one restaurant in X City on TripAdvisor. And they wind up up just doing really well with a bunch of tourists basically.
- [28:12]
A
That's right. It's optimization depending upon how you view your channels. And I think you can do it on the grubhubs too. You just have to be super cognizant of the cost. Right? If Every grubhub customer is going to effectively cost you a 30% surcharge. Then you need to figure out how do I get that customer onto my site? How do I cover the cost of that? 30% is coming out of my marketing budget for the first order, do I mark up the price of my food 30%? So these are all things that the restaurants have to figure out and it's non trivial.
- [28:41]
B
So are you like the big picture then? Are you pessimistic about the prognosis for a mom and pop restaurant? Like there's so many different moving parts that they need to be cognizant of that ultimately these big enterprises with multiple storefronts and chain franchises and brands are always just going to keep on growing. Or do we romantics still have a little chance?
- [29:07]
A
Well, every one of those large chains you talked about started as a mom and pop. I don't think you extinguish that part of the market ever. There's always going to be entrepreneurs out there looking to build new concepts and try new things. And some of them might be big and other ones might fizzle out.
- [29:23]
B
I don't want my local restaurant to be an entrepreneur with a concept. I want my local restaurant to be. Be a cook with food. But that's. That's just me. Jordan Baylor, thank you so much for joining us. We really, really appreciate it.
- [29:37]
A
Thanks fellow.