Well, not everyone. This is mostly limited to the major chains.
But by "scary", I refer mostly to the outright purchase of a data mining company, and they're already talking about intrusive surveillance such as using license plate readers.
They're also being pretty up-front about wanting to use this to straight-up manipulate people. I don't know whether to give them credit for honesty or not (since all the cool kids are manipulating people these days), but it makes me recoil nonetheless.
they're already talking about intrusive surveillance such as using license plate readers.
If you drive onto their property, and McDonald's Corporation is almost always the landlord of a franchised store, are they really intruding on your privacy?
Yes, clearly. The question I think you're really getting at is "have I given consent to that?" And the answer to that is, assuming that I have been notified of the practice before I enter their property, yes.
Which is precisely why these systems will mean that I won't go onto their property.
I think customers will find this very annoying. There's already way too many options on the menu. And their in store self service ordering kiosks are horrendously complicated already.
They should eliminate about 60% of the menu and focus on speed and quality for the remainder.
When I go to McDonald's, if they suggest anything that I didn't ask for it pisses me off. Their food is pretty shit at best. The only reason I go there is to get something quick without any ceremony. If they can't do that, there's no reason for me to even go.
That's the idea. Part of what they're trying to do is speed up drive-thru service, since that's now where the majority of customers visit.
Humans can't predict what to make in the minutes before a customer pulls up to the speaker, but an ML tied into a number of systems and inputs might be able to do a better job.
It's not really anything new, to be honest. Back in the days where restaurants cooked and wrapped the burgers ahead of time and held them in a warming bin for 10 minutes at a time (at least that's what the spec dictated) the managers would keep detailed records on customer count, weather, cars in the lot, local events, etc and try to make predictions based on historical data. Wasn't the most effective, but it helped. This is just a modernization of that idea.
Man was it fast when you could come in and they had burgers in the bin waiting. They got shamed into cooking to order, and the downside is waiting; I'm never ordering a quarter pounder again since they changed those recently to take forever to cook.
From what I know, speeding up the whole "made to order" menu is one of the prime motivators of all this technology. The fresh burgers are a leap ahead in quality but an order of magnitude slower to make.
MCD will most likely spend half a billion dollars to get that burger on the grill 90 seconds earlier if it's possible.
No the burgers are not a leap ahead. They are marginally better, but that isn't saying much. I find them barely edible but tolerate them if I'm pressed for time.
> Humans can't predict what to make in the minutes before a customer pulls up to the speaker, but an ML tied into a number of systems and inputs might be able to do a better job.
Mcdonalds has a mobile app, where you can order and your food will be ready before you arrive. Wouldn't it have been better to push customers towards this channel, where they can order what they want instead of ML guessing, and saving a couple hundred million dollars along the way?
As others have mentioned, Dominoes doesn't use ML/AI, but you can order a pizza crazy fast through the app and know when it's done (and that seems to be enough).
I don't know if I could draw an analogy to pizza here. It's a different type of purchase. You know it's going to take 15-20 minutes to make and more often than not it's going to be delivered. It also survives the delivery/pickup process with better quality. Have you ever eaten MCD french fries that were more than 5 minutes old?
From my interactions with their people, the mobile app is the conduit to the future but there are a lot of kinks still left to be worked out.
For example: you order on the app, but when exactly do you start the food? You might be 30 seconds from the store or 30 minutes. The initial idea was to geofence the store and trigger the kitchen when you're on the way. Perhaps that's still going, but maybe the ML is an assist on that data. You also get a ton of marketing and demographic information out of that. There are also the usual privacy concerns.
Remember that there's also a ton of data being generated in the other direction. Predictive analytics on the equipment (everything in the kitchen will be an edge node in a few years), POS/cash accounting, real-time sales data, store traffic, crew work history, etc.
They're also experimenting with dynamic pricing on food, like it or not.
Maybe it will allow them to simplify the menu. Also, they have a case-study on their work with Hello Fresh that talks about how they worked to acquire, basically, unique ID for customers (not sure why that should have been any technical challenge, but anyway...) If that is an initial focus & precursor to their services, I can imagine that they will capture license plate numbers and associate them with each order. Then, the next time you pull up, you're shown a customized (and simplified) menu based on your likely preferences.
> And their in store self service ordering kiosks are horrendously complicated already.
Small rant: a weird pleasure in Japan is using restaurant touchscreen consoles that don't take several million very laggy taps to complete a simple order. Places like Yayoiken[1] and Matsuya[2] have this down to a science. It's partly because of fewer menu options (although it's not like either one has a miniscule menu[3]).
Insert cash, tap tile, ticket immediately dispenses. Tap change button when done. The combination of denser screen layouts, decently responsive touchscreens and a near-complete lack of interstitial animations makes using these things extremely snappy. If you know what you want, the entire process can take maybe ten seconds.
It seems like when your company's revenue has been flat for over a decade, you should focus on improving your core offering of, you know, food quality, and not find new ways to annoy customers with data mining and extreme personalization.
In 2009 Dominos realized their food sucked, committed to improving it, and have since been doing very well. (Their stock has more than quadrupled.) At some point McDonalds will need to do the same.
I remember the World of Warcraft add-on that made news headlines because it allowed you to order a pizza in-game, during a time when we were seeing constant stories about someone dying (or causing others to die) thanks to gaming addiction and forgetting to sleep/eat/feed children.
from what I can see the video showed that machine separates ripe grapes in good color and shape from the rest of the mix. did you see the machine taste wine and go back to tune up its sorting algorithm?:)
I'm no domain expert but there's graphs for all sorts of things when it comes to food (usually used for quality control). You can probably just run computer vision over said graphs and get somewhere reasonable.
I actually haven't seen any statistics on this, but based on knowing how the industry works in the U.S. it seems like if it's not already the majority it's going to be the majority very soon.
Basically there are thousands of vineyard owners and thousands of wineries, but most of the vineyard owners take their grapes to the same handful of places to process them in between when they're picked and when they're sent off to the wineries. And these machines can process enormous volume, and there is tons of risk around labor.
And especially given that the majority of wine is made by the same handful of companies and sells for less than $10, I can't see any way that the majority of grapes aren't passing through these things at this point even if it's not really talked about.
Color sorting of fruits and food grains is definitely a pre-machine learning problem. On the other hand, there are very problems which cannot be improved using machine learning. Incidentally, my father ran a startup building color sorting machines for coffee beans with analog electronics ~25 years ago.
As another commenter pointed out, Dominos is kinda the poster child for fast food companies driving growth by embracing technology. They started investing in online ordering back in 2007, and their stock has gone up more than 20x since then.
And if you look at how Dominos describes itself, they're clearly focused on technology (or at least want you to think they are). For example, half of their standard investor blurb is about technology:
> Emphasis on technology innovation helped Domino's achieve more than half of all global retail sales in 2017 from digital channels, primarily online ordering and mobile applications. In the U.S., Domino's generates over 60% of sales via digital channels and has produced several innovative ordering platforms, including Google Home, Facebook Messenger, Apple Watch, Amazon Echo, Twitter and text message using a pizza emoji. In late 2017, Domino's began an industry-first test of self-driving vehicle delivery with Ford Motor Company – and in April 2018, launched Domino's HotSpots™, featuring over 200,000 non-traditional delivery locations including parks, beaches, local landmarks and other unique gathering spots.
> Domino's generates over 60% of sales via digital channels
This blows my mind. I remember when the online pizza tracker first came out, and I was blown away. I couldn't imagine the logistical nightmare they must have solved to get this installed in 5,000+ stores in the U.S. It really was ahead of it's time, and I can't think of any other online food ordering platform that gets this close.
It was a nightmare. We all had a deadline to book these installers for the Cisco VPNs and VM servers. IIRC IBM wrote the code, but Domino's retained the rights.
It was a great time for the company. Total 180 on quality, investing heavily in the right tech. 3 years after we installed the server & thin clients all around, 33% of orders and 50% of revenue aas online.
Online sales drove order frequency, ticket price and customer satisfaction while lowering costs. It was such a genius move.
Source: I was a Domino's GM and franchise for 17 years and saw this transition.
It's kind of interesting how Little Caesars found success with the opposite strategy. Instead of "our food sucks, lets make it better and improve the ordering experience too", they said "I bet people won't mind if their pizza tastes like cardboard as long as it only costs $5, they don't have to order it, and can pick it up on the way home".
Little Caesars has an app! Never tried it myself but they have a little cage where you scan your phone (or whatever) and the door opens to get your pizza.
I usually just call ahead instead of having to sit around 10-15 minutes or ask what's ready if I don't want something specific.
Disclaimer: I eat Little Caesars probably way too much.
Nothing to do with the service guarentee^H^H^H^H^H^H^H^H^H "commitment".
It was a redesign from the ground up of the pizza. New dough & sauce formula. A butter/garlic/seasoning added to the crust.
Also Quantity of cheese on a plain pie increased by 50%. A 14" pie went from 7oz to 10oz. Owners lost their shit over this mandated cost increase. Everyone shut up about 4 months later once sales were up 50%.
And a very, very nice advertising campaign. The CEO got on TV and said "Our pizza sucks. Sorry. We know, we listened, we fixed it. Buy three of them for $15 and if you don't like it, we will refund your money no questions asked".
The new customer satisfaction rules were another source of contention. The 100% satisfaction guarentee made franchisees, especially those in.... "urban" environments very nervous. They thought there would be a line out the door of people scamming. My store in such a "low-income, high-population density" environment.
Corporate came down hard on them. Owners, even huge multi-store franchises with 10s of millions in sales were told where the door was if they didn't like it. I was very proud of the central office. They took what could be an excuse to make two rules - One for Flatbush Avenue and another for Newport Beach - and applied it across the company. It really endeared higher management to the workers, many of which lived in just such areas.
I was never worried, and it turned out to not be a problem. The number people who asked for a refund for questionable reasons were very very small and an easily absorbed cost of business.
How many sales are through there own digital channels these days? Around me they've had to go on uber eats, menulog and sites like that to stop hemorrhaging customers to stores with no online presence that they didn't have to compete with a few years ago and they're no longer cheaper compared to much better stores.
McDonald's is a massive bottom-line franchise, so enforcing new standards that cut into margin and getting paying franchisees to put their stores on board is a recipe for disaster. I wish McDonald's would improve their quality too but that is just not what McDonald's is.
Dominos has been doing well thanks to early adoption of tech and using it to integrate with customers in new ways that the competition hasn't. They proved the success in fast-food franchises with this, and now McDonald's is trying to differentiate with tech too with this acquisition.
What quality improvement do you think McDonald's needs? They have already gone to excessive lengths to appeal to parents and advocacy groups by adding healthy meal options (apple slices, milk, salads) and reducing portion sizes (which helps them more in the profit department, than it helps the overweight consumers) and clearly showing the breakdown of each item on a large prominent poster near the register. They're a fast food restaurant who has already gone above and beyond all of their closest competitors IMO.
Offering packaged apple slices and milk is going to excessive lengths? Really? That seems closer to the bare minimum than “above and beyond” but maybe you have difference standards than me.
Excessive lengths would be McDonalds installing cold-pressed juice bars. An above and beyond McDonalds would just be your average In-N-Out.
lol people who harp about the superiority of in n out are so blasse in taste. its the same shit, no need to market their burgers for them in your own head.
Ya, I lived in LA for a year and didn't understand the In n Out hype. Pretty middle of the road every time I had it. Dick's in Seattle is better in my opinion
If you can’t taste or smell the difference between McDonalds and In-N-Out then you must have poor taste and senses. The meat, buns, and fries are completely different. I’m not saying it’s the best burger in the world, but it shows you can get better quality burgers than what McDonalds is selling for about the same price. McDonalds chooses not to care about quality.
That’s just a choice though. In-N-Out is cheap but the food is high quality. McDonalds would rather use marketing to compensate for quality. It comes down to priorities.
In-N-Out isn't a franchise. It's a private, family owned business that operates each individual store, so they have the control to enforce quality and employment standards while maintaining margins.
It hasn't quite been 10 years. They had significant revenue growth after 2009 [0] And annual gross profit is the highest it's been since 2013. [1]
In the last 2 years they have seen significant revenue decrease, but primarily because they sold off a significant number of stores to franchisees, which transfers that revenue off their books [2]
That said your underlying point, that they face "headwinds" in the market right now, is correct. Competition from "fast casual" dining has eroded growth that might otherwise have gone to "quick service" outlets like McDonalds, and growth in the industry continues to strongest in the fast casual segment [3]
They revamped their quarter pounders a ~year ago to be much fresher, tastier, juicier burgers. They serve more gourmet-style toppings now and a variety of cheeses. They're experimenting with new items like the morning "donut sticks" which are surprisingly good. The buttermilk chicken tenders are worlds beyond the old nuggets. And you can get breakfast all day, with eggs that are now fresh-cracked.
They still sell all their old stuff because people still love it. But if you think their menu hasn't been increasing in quality for those who want it, you haven't been paying attention. Some people prefer not to believe it.
But it's still McDonald's. People want and like McDonald's. They're not looking for fine dining.
McDonalds as a company has fine food quality based on the fresh hot and delicious McDonalds I got in Japan. I believe McDonalds US has a store management and personnel problem, i.e. that no one gives a shit. Even after the “fresher” Quarter Pounders the meat was warmer but I sometimes got buns so dry and stale they were falling apart in the box. Sure theres lots of timers and tech to make sure workers can’t burn the fries but it appears to me no one in charge of overall quality or consistency.
That's what happens when you have a culture that so heavily looks down on people who work these kinds of jobs. Failures who couldn't make it in school work at Mcdonalds they say, and failures they'll get, making their burgers.
> It seems like when your company's revenue has been flat for over a decade, you should focus on improving your core offering of, you know, food quality, and not find new ways to annoy customers with data mining and extreme personalization.
That's wrong. McDonalds is real estate holding company that happened to sell fast food.
And why do you think that real estate is so valuable? Why do they insist on blanketing all their great real estate with McDonald's restaurants? Perhaps because it's a great place to sell McDonald's food?
Because those typically are newer buildings accessible to cars with drive ins. Those buildings are very valuable.
Contrary to strange beliefs of HN, most of business is really simple: make a little money every day over a long period of time. Invest in real estate. 50 years later it would be a very good very viable business.
Same goes for lawyer offices ( not law firms ). The "rich" lawyers do very well by having a small practice and buying a building that they use for law offices. Upon retirement most of their money comes not from selling the book of business to a different attorney but from selling a building that they have paid off.
> It seems like when your company's revenue has been flat for over a decade, you should focus on improving your core offering of, you know, food quality
Quality is subjective, and analytics is a tool for hitting the target the market is looking for.
Why did you pick a company that is famous for claiming they are about tech not food?
'A Tech Company That Sells Pizza' Delivered Huge For Domino's'
'Domino’s Pizza CEO Patrick Doyle has been talking about being a tech company that sells pizzas for years.'
Interesting that they decided to acquire the company rather than just become a customer. Perhaps they saw strategic value of depriving other potential customers? In any case, the article was light on specific insights McDonalds has received and instead gave hypotheticals, so I thought I'd share this from Dynamic Yield-- it's a case study with more specifics so it gave me a much better idea of what they do:
Dynamic Yield (reasonably) seems to focus initial efforts on associating a unique id with customers. Given that, it's not much of a leap at all to think that McDonald's will start capturing license plates in the drive through and associating them with the correct order. (Heck, they might do it already) With Dyanamic Yield they then offer custom recommendations the next time that plate number is detected.
I'm not sure if that creeps me out, or if the data geek in me likes the elegant solution to a problem. Why choose? It can be both.
Hey, you're right. And my bartender always knows I want "the usual"! Just because the cookie exists in his neural structure doesn't make it any less of a violation of the GDPR :)
I don't think you can really compare the implicit "tracking" that occurs with face-to-face interaction with tracking through technology - there's a huge difference in scale. It would be like saying there's no difference between mom and pop shops and Walmart because they both just sell commodities.
"If someone orders two Happy Meals at 5 o’clock, for instance, that’s probably a parent ordering for their kids; highlight a coffee or snack for them, and they might decide to treat themselves to a pick-me-up. And as with any machine-learning system, the real benefits will likely come from the unexpected."
It looks like 'machine learning', 'data mining', etc. gets thrown around a lot even when irrelevant. That machine was never going to find out that it was a parent in there ordering 2 happy meals for their kids. Its more like 'if 2 an item tagged for_kids sold then suggest item tagged for_adults, if weather_temp > 40 degreea then suggest ice coffee otherwise hot coffee' where is machine learning in that?
I honestly would prefer they showed me my recent orders based on my cars license plate just like dominos.com does when I go back on their website
- Presumably you can get 90% of the value for 5% of the price. McDonald's isn't Amazon. They have 100 (?) menu items, not 1,000,000 products. Why are they spending $300M instead of hiring a few data scientists? Or just be a customer of the company?
- Dynamic Yield has many other clients, and it sounds like they'll continue to serve those. Is McDonald's the right owner for a high-tech customer intelligence product? It just feels like such a bizarre cultural and core-competency fit. Maybe invest partly in it, but buy it outright? I don't get it. Seems like you'd rather share the board with other experts?
I think you are drastically underestimating McDonald's - they have 36k+ stores in almost every country on the planet serving 69MM+ customers daily with a brand recognition in line with Apple, Coca Cola, and Nike. They also spend money on advertising across every conceivable channel.
Number of SKUs has nothing to do with it (which likely numbers in the many hundreds across the globe considering regional & seasonal specialities).
McDonald's has everything to gain by infusing this kind of DNA into their operations.
> McDonald's has everything to gain by infusing this kind of DNA into their operations.
That seems logical but how much will it infuse Machine Learning into McD's DNA if the acquired startup sits in Israel? It seems to me the employees will not even interact much with each other, how can there be much of incorporating thinking, methods or practices?
> It seems to me the employees will not even interact much with each other, how can there be much of incorporating thinking, methods or practices?
What does physical proximity have to do with anything?
McDonalds does $20B+ in revenues, has 210k direct employees and 1.9M when you include franchises. At this scale, the only people that need to interface is senior management.
Common everywhere, they're "grill slips." It's not a new SKU, but a customization to it. Still, wonder how much analytics they pull on that when designing new products.
This has been the secret for decades in most fast food restaurants to make sure your order is made fresh (by making requests that ensure a pre-cooked item can't be used).
It’s also extremely common that a lot of the chains will have a “fuck you” mentality and just reuse an existing burger any way if they know you’re just trying to skirt the freshness. The real pro tip is to just ask nicely for a fresh burger, they’ll do an Ask Me, and everyone is happy.
Source: 2 years at a McDonald’s in my younger days.
EDIT: That is to say this isn't really a secret, it happens all the time. But when you're working with teenagers working for minimum wage, who hate when others try to be sneaky/talk down on them just being upfront and nicely asking what you want, and being sure to say "I know it'll take a bit longer, that's fine." is better than the alternative.
On this note, and current McDonald's employees are going to hate me... but if you want fresh, hot and crispy fries, ask for "no salt". They'll have to put a new batch down, clean off the scooper and then give you fresh out of the frier fries. Then ask for a salt packet, because they're shit without salt.
Just ask for a fresh batch, they'll do an Ask Me (same as before) and they aren't nearly as annoyed if you let them know you're aware it'll take longer.
Ya gotta say. I've never worked at McDonalds, but I worked at A&W a long time ago, and there was no problem when someone wanted somethinf fresh made. The burgers took like 2 minutes to cook anyway. They sell fried chicken there though, that was kind of a pain to cook up fresh, it took 20 minutes to fry the chicken.
On that note, I used to work the morning shift, but i remember one day, for some reason, nobody showed up for the afternoon shift so I had to stay. I had no idea I was supposed to start cooking chicken around 4 or so. The families, or people out to get an easy meal for their families, started ordering chicken. So many buckets of chicken...There was not enough chicken by far. People began to grow impatient. Around the time I finally managed to get on top of the chicken orders, people finally showed up for work and I was free.
Sorry, I was originally only going to write the first part of the comment, but as I wrote it, that day came to mind. I'd forgotten about that.
An Ask Me is a selection on the screen for an option that can’t be added as a customization. It’s to “ask the counter” for what is wanted. Like a bunless burger or what have you.
> Presumably you can get 90% of the value for 5% of the price. McDonald's isn't Amazon.
"McDonald's isn't Amazon" seems like a good enough reason.
Amazon can "just hire a few PhDs/seasoned engineers", plug them into the right places in the company, and expect good things to happen.
McDonald's would need to hire a new team from the leadership down and would be working up-wind in a hot seller's labor market at every level.
> They have 100 (?) menu items, not 1,000,000 products.
McDonald's is a huge company with lots of opportunities to apply ML. Store operation, hiring, supply chain optimization, menu/product design, lots of resource scheduling problems, marketing, etc.
I can easily see McDonald's making back hundreds of millions very quickly just by decreasing the friction and amount of time required to make improvements in these and other areas.
> Is McDonald's the right owner for a high-tech customer intelligence product? It just feels like such a bizarre cultural and core-competency fit.
Automated ordering is becoming a core part of McDonald's customer experience. Automated and augmented decision making is becoming a core part of McDonald's franchise management. Outsourcing those things would mean putting the fate of the company's future in someone else's hands.
Plus, there's huge upside to owning that IP. McDonald's is taking leadership on automating away the sales clerk and decreasing reliance on skilled/knowledgeable franchise owners and managers.
I know from experience that you can put together a very good full-suite team of 20 people in SFBA with a 5-year budget in the tens of millions. That's a fraction of a $300M acquisition (though obviously the acquisition brings IP, relationships, a team that already works together, etc.)
Such a team is on the small side for a Fortune 500, but definitely large enough to do real work and more than pay for themselves (especially if you're introducing serious ML into your company for the first time).
I don't know about McD's, but I work at another very large franchised quick-service restaurant (QSR) (not McD's, but in a different category).
For companies like these, that operate in 10's of thousands of franchisee-owned locations, the number of products and the combination of configurations is not merely a function of the things you see on the menu. It's all of those products, with their different combination of parts (beef patties, lettuce, etc.), and then franchisee and regional variations (In some countries, you can't tell a franchisee what they can or can't sell, etc.)
Add on top of that customer modifications to the product in their order (extra pickles, no onions, etc.)
Why does this matter? It drives lots of stuff - food costs, inventory, inventory & sales forecasting (how many pickles do I need this week?) new item research, profit margins, etc.
So you take this, multiply it by 10,000, 15,000, or, in McD's case, 36,000 stores across 100+ countries (half both those numbers in my company's case) and you're talking about vast amounts of information across millions of transactions every day, and that is in fact Big Data.
Whether this is a smart move, I'm not sure. I'm of the opinion that if a piece of technology is core to the _operation_ of your business (for a QSR, things like POS systems, backend administrative systems, etc.), it should be in your house. Other things (HR systems, big data systems) should not be. But maybe McD's sees it differently.
Domino's has even less menu choice than McDonalds... But you do make a good point. I went into the flagship mcd's in Chicago the other day and those touchscreens, while not tinesaving at all, do allow you to get deep into a big menu and make all kinds of modifications. My friend accidentally ordered a bunless burger. As for me, when it said mcgriddles were sold out/greyed out, I simply ordered a sausage biscuit but subbed mcgriddle for the biscuit (the counter guy later told me they were still sold out and offered me a free mcdouble instead).
McDonalds uses this type of data in so many ways. When I worked for them, they uncovered a franchisee defrauding them and the IRS. This was when x86 POS systems were first replacing Scantron menu systems. This franchisee hired someone to customize the POS system to underreport sales. This affected both the tax collections, as well as franchise fees due. How was he found? Well, all food/paper sales were through one vendor that reports sales to the mothership. They could see that he was ordering enough supplies for X, while reporting Y sales. He skimmed about 10% extra over two years before they busted him. So for roughly $200K he lost all four franchises, and faced a huge tax issue.
This type of data processing across all the McOpCo stores and franchises will be invaluable.
The kind of items that McDonalds has in India is very very different from what I have seen in the US. While I agree with your sentiment, you are also vastly underestimating the needs of McDonalds.
It's mentioned in the article, but much of what drives McDonald's profitability is their supply chain and logistics, and r&d around food prep/meal production. There are huge dollar volumes there at very low margins and it is an area of great advantage (not only in cost, but in what it's possible to do product-wise) if done well. It's a perfect area for data science / ML type applications.
It depends how you view McDonald's. If you think it's the biggest employer of burger flippers in the world, then you're probably right... Seems like a weird fit.
But if you think of them as experts in agriculture, farming, supply chain and logistics, kitchen automation, marketing, global scale business build out... Then I could see how an acquisition like this could be synergistic.
Don't forget that the Trump tax cuts have provided big companies with a windfall. They have way more free cash flow than they did just 2 years ago. So they also are desperate to find new ways to deploy cash.
The world's biggest employer of burger flippers also has a clear economic interest in developing tech that'll be the world's biggest unemployer of burger flippers.
McDonalds is heavily investing into automation for the ordering process; and as soon as it becomes technically practical and reliable enough, they're going to want to automate even more.
As everybody likely knows, a significant part of McDonald's's revenue is operations with realty, which they have plenty. I wonder what else are they leveraging from their position of a huge company with a truly global presence.
I mean, Amazon also sells books as it used to, and selling stuff is their core business, but they are not limited to that.
It might be something like what FaceBook did with the VPN app that got them into trouble: if they continue to serve the other customers, they get to see their data -- what's trending in the market, etc.
I do worry about the loss of jobs at this level though.
Is it right to think that $300M investment + time to turn it into usable product for the restaurants == Loss of many minimum wage (or slightly more) jobs to make back that $300M
My local Wal*Mart now has 25 self checkouts and only 2 people manned registers.
It might be great news on some level, but that many jobs still effects the GDP and more importantly people's lives until they find other work, if they can, and it may bot be easy for them. Enjoy it or not. Putting food on the table and the next day not being able to is hard on anyone. Multiply this my potentially millions of jobs.
Travis from UBER is starting a kitchen sharing company.
McDonalds has been in the real estate game within hospitality larger and longer than anyone.
This move makes a lot of sense and hopefully they move smartly to compete as the industry evolves.
McDonald’s can’t scale ethically raised food to their restaurants at the price points that matter to their customers but they can bring operational and supply chain expertise to startup restaurants.
This acquisition is a prime example of what the future of food is going to look like. It’s interesting how food has been one of the last major consumer categories to be eaten by software. But it’s finally happening!
Three simultaneous changes are happening related to the availability, personalization, and niche focus of food. I’ve done a deep dive on this (1) if you want to read more.
128 comments
[ 4.9 ms ] story [ 190 ms ] threadBut by "scary", I refer mostly to the outright purchase of a data mining company, and they're already talking about intrusive surveillance such as using license plate readers.
They're also being pretty up-front about wanting to use this to straight-up manipulate people. I don't know whether to give them credit for honesty or not (since all the cool kids are manipulating people these days), but it makes me recoil nonetheless.
If you drive onto their property, and McDonald's Corporation is almost always the landlord of a franchised store, are they really intruding on your privacy?
Which is precisely why these systems will mean that I won't go onto their property.
And as a result, everyone gets hotter fries?
I don't see why having Math help make better decisions is a bad thing.
It isn't. But collecting unnecessary data and using it to manipulate people certainly is.
They should eliminate about 60% of the menu and focus on speed and quality for the remainder.
If I know facebook, they are going to make McDonalds addicting AF
That's the idea. Part of what they're trying to do is speed up drive-thru service, since that's now where the majority of customers visit.
Humans can't predict what to make in the minutes before a customer pulls up to the speaker, but an ML tied into a number of systems and inputs might be able to do a better job.
It's not really anything new, to be honest. Back in the days where restaurants cooked and wrapped the burgers ahead of time and held them in a warming bin for 10 minutes at a time (at least that's what the spec dictated) the managers would keep detailed records on customer count, weather, cars in the lot, local events, etc and try to make predictions based on historical data. Wasn't the most effective, but it helped. This is just a modernization of that idea.
MCD will most likely spend half a billion dollars to get that burger on the grill 90 seconds earlier if it's possible.
Mcdonalds has a mobile app, where you can order and your food will be ready before you arrive. Wouldn't it have been better to push customers towards this channel, where they can order what they want instead of ML guessing, and saving a couple hundred million dollars along the way?
As others have mentioned, Dominoes doesn't use ML/AI, but you can order a pizza crazy fast through the app and know when it's done (and that seems to be enough).
From my interactions with their people, the mobile app is the conduit to the future but there are a lot of kinks still left to be worked out.
For example: you order on the app, but when exactly do you start the food? You might be 30 seconds from the store or 30 minutes. The initial idea was to geofence the store and trigger the kitchen when you're on the way. Perhaps that's still going, but maybe the ML is an assist on that data. You also get a ton of marketing and demographic information out of that. There are also the usual privacy concerns.
Remember that there's also a ton of data being generated in the other direction. Predictive analytics on the equipment (everything in the kitchen will be an edge node in a few years), POS/cash accounting, real-time sales data, store traffic, crew work history, etc.
They're also experimenting with dynamic pricing on food, like it or not.
Small rant: a weird pleasure in Japan is using restaurant touchscreen consoles that don't take several million very laggy taps to complete a simple order. Places like Yayoiken[1] and Matsuya[2] have this down to a science. It's partly because of fewer menu options (although it's not like either one has a miniscule menu[3]).
Insert cash, tap tile, ticket immediately dispenses. Tap change button when done. The combination of denser screen layouts, decently responsive touchscreens and a near-complete lack of interstitial animations makes using these things extremely snappy. If you know what you want, the entire process can take maybe ten seconds.
[1] https://s3-media1.fl.yelpcdn.com/bphoto/PzkepdRwWT6NaE4XDgKz...
[2] http://3.bp.blogspot.com/-cJ8-6hscmOQ/ViHzR4f0EmI/AAAAAAAABc...
[3] https://www.matsuyafoods.co.jp/menu/
In 2009 Dominos realized their food sucked, committed to improving it, and have since been doing very well. (Their stock has more than quadrupled.) At some point McDonalds will need to do the same.
https://www.forbes.com/sites/aliciakelso/2018/04/30/delivery...
Yup, I remembered correctly! https://boingboing.net/2005/02/18/everquest-now-with-p.html
https://www.youtube.com/watch?v=aY4KPMYd4xw
Where are you pulling this number from?
Basically there are thousands of vineyard owners and thousands of wineries, but most of the vineyard owners take their grapes to the same handful of places to process them in between when they're picked and when they're sent off to the wineries. And these machines can process enormous volume, and there is tons of risk around labor.
And especially given that the majority of wine is made by the same handful of companies and sells for less than $10, I can't see any way that the majority of grapes aren't passing through these things at this point even if it's not really talked about.
And if you look at how Dominos describes itself, they're clearly focused on technology (or at least want you to think they are). For example, half of their standard investor blurb is about technology:
> Emphasis on technology innovation helped Domino's achieve more than half of all global retail sales in 2017 from digital channels, primarily online ordering and mobile applications. In the U.S., Domino's generates over 60% of sales via digital channels and has produced several innovative ordering platforms, including Google Home, Facebook Messenger, Apple Watch, Amazon Echo, Twitter and text message using a pizza emoji. In late 2017, Domino's began an industry-first test of self-driving vehicle delivery with Ford Motor Company – and in April 2018, launched Domino's HotSpots™, featuring over 200,000 non-traditional delivery locations including parks, beaches, local landmarks and other unique gathering spots.
https://www.prnewswire.com/news-releases/dominos-pizza-annou...
This blows my mind. I remember when the online pizza tracker first came out, and I was blown away. I couldn't imagine the logistical nightmare they must have solved to get this installed in 5,000+ stores in the U.S. It really was ahead of it's time, and I can't think of any other online food ordering platform that gets this close.
It was a great time for the company. Total 180 on quality, investing heavily in the right tech. 3 years after we installed the server & thin clients all around, 33% of orders and 50% of revenue aas online.
Online sales drove order frequency, ticket price and customer satisfaction while lowering costs. It was such a genius move.
Source: I was a Domino's GM and franchise for 17 years and saw this transition.
I usually just call ahead instead of having to sit around 10-15 minutes or ask what's ready if I don't want something specific.
Disclaimer: I eat Little Caesars probably way too much.
Is this alluding to "30 minutes or less"?
It was a redesign from the ground up of the pizza. New dough & sauce formula. A butter/garlic/seasoning added to the crust.
Also Quantity of cheese on a plain pie increased by 50%. A 14" pie went from 7oz to 10oz. Owners lost their shit over this mandated cost increase. Everyone shut up about 4 months later once sales were up 50%.
And a very, very nice advertising campaign. The CEO got on TV and said "Our pizza sucks. Sorry. We know, we listened, we fixed it. Buy three of them for $15 and if you don't like it, we will refund your money no questions asked".
The new customer satisfaction rules were another source of contention. The 100% satisfaction guarentee made franchisees, especially those in.... "urban" environments very nervous. They thought there would be a line out the door of people scamming. My store in such a "low-income, high-population density" environment.
Corporate came down hard on them. Owners, even huge multi-store franchises with 10s of millions in sales were told where the door was if they didn't like it. I was very proud of the central office. They took what could be an excuse to make two rules - One for Flatbush Avenue and another for Newport Beach - and applied it across the company. It really endeared higher management to the workers, many of which lived in just such areas.
I was never worried, and it turned out to not be a problem. The number people who asked for a refund for questionable reasons were very very small and an easily absorbed cost of business.
Dominos has been doing well thanks to early adoption of tech and using it to integrate with customers in new ways that the competition hasn't. They proved the success in fast-food franchises with this, and now McDonald's is trying to differentiate with tech too with this acquisition.
Excessive lengths would be McDonalds installing cold-pressed juice bars. An above and beyond McDonalds would just be your average In-N-Out.
If you can’t taste or smell the difference between McDonalds and In-N-Out then you must have poor taste and senses. The meat, buns, and fries are completely different. I’m not saying it’s the best burger in the world, but it shows you can get better quality burgers than what McDonalds is selling for about the same price. McDonalds chooses not to care about quality.
In the last 2 years they have seen significant revenue decrease, but primarily because they sold off a significant number of stores to franchisees, which transfers that revenue off their books [2]
That said your underlying point, that they face "headwinds" in the market right now, is correct. Competition from "fast casual" dining has eroded growth that might otherwise have gone to "quick service" outlets like McDonalds, and growth in the industry continues to strongest in the fast casual segment [3]
[0]https://www.macrotrends.net/stocks/charts/MCD/mcdonalds/reve...
[1] https://www.macrotrends.net/stocks/charts/MCD/mcdonalds/gros...
[2] https://www.marketwatch.com/story/mcdonalds-revenue-hit-by-f...
[3] https://www.restaurantbusinessonline.com/financing/technomic...
They revamped their quarter pounders a ~year ago to be much fresher, tastier, juicier burgers. They serve more gourmet-style toppings now and a variety of cheeses. They're experimenting with new items like the morning "donut sticks" which are surprisingly good. The buttermilk chicken tenders are worlds beyond the old nuggets. And you can get breakfast all day, with eggs that are now fresh-cracked.
They still sell all their old stuff because people still love it. But if you think their menu hasn't been increasing in quality for those who want it, you haven't been paying attention. Some people prefer not to believe it.
But it's still McDonald's. People want and like McDonald's. They're not looking for fine dining.
In Japan. A country with seriously incredible food.
Gets McDonalds.
Wow.
That's wrong. McDonalds is real estate holding company that happened to sell fast food.
https://qz.com/965779/mcdonalds-isnt-really-a-fast-food-chai...
Contrary to strange beliefs of HN, most of business is really simple: make a little money every day over a long period of time. Invest in real estate. 50 years later it would be a very good very viable business.
Same goes for lawyer offices ( not law firms ). The "rich" lawyers do very well by having a small practice and buying a building that they use for law offices. Upon retirement most of their money comes not from selling the book of business to a different attorney but from selling a building that they have paid off.
Quality is subjective, and analytics is a tool for hitting the target the market is looking for.
I never thought it was all that good, and haven’t eaten it in a decade.
https://www.insidescandinavianbusiness.com/article.php?id=34...
'A Tech Company That Sells Pizza' Delivered Huge For Domino's' 'Domino’s Pizza CEO Patrick Doyle has been talking about being a tech company that sells pizzas for years.'
https://www.forbes.com/sites/aliciakelso/2018/04/30/delivery...
https://www.dynamicyield.com/case-studies/hellofresh/
I'm not sure if that creeps me out, or if the data geek in me likes the elegant solution to a problem. Why choose? It can be both.
Feature: for_kids, for_adults
Machine learning allows you to generalize inference across a large number of features.
Instead of hardcodong control flow, you train to maximize an objective function.
The belief is that once trained, you can infer cheaply and scalably.
- Presumably you can get 90% of the value for 5% of the price. McDonald's isn't Amazon. They have 100 (?) menu items, not 1,000,000 products. Why are they spending $300M instead of hiring a few data scientists? Or just be a customer of the company?
- Dynamic Yield has many other clients, and it sounds like they'll continue to serve those. Is McDonald's the right owner for a high-tech customer intelligence product? It just feels like such a bizarre cultural and core-competency fit. Maybe invest partly in it, but buy it outright? I don't get it. Seems like you'd rather share the board with other experts?
Number of SKUs has nothing to do with it (which likely numbers in the many hundreds across the globe considering regional & seasonal specialities).
McDonald's has everything to gain by infusing this kind of DNA into their operations.
That seems logical but how much will it infuse Machine Learning into McD's DNA if the acquired startup sits in Israel? It seems to me the employees will not even interact much with each other, how can there be much of incorporating thinking, methods or practices?
What does physical proximity have to do with anything?
McDonalds does $20B+ in revenues, has 210k direct employees and 1.9M when you include franchises. At this scale, the only people that need to interface is senior management.
https://en.wikipedia.org/wiki/Ray_Kroc
Even its title, similar to Kroc (his: Grind It Out).
Yes, there was a time when Starbucks was not doing well.
Source: 2 years at a McDonald’s in my younger days.
EDIT: That is to say this isn't really a secret, it happens all the time. But when you're working with teenagers working for minimum wage, who hate when others try to be sneaky/talk down on them just being upfront and nicely asking what you want, and being sure to say "I know it'll take a bit longer, that's fine." is better than the alternative.
On that note, I used to work the morning shift, but i remember one day, for some reason, nobody showed up for the afternoon shift so I had to stay. I had no idea I was supposed to start cooking chicken around 4 or so. The families, or people out to get an easy meal for their families, started ordering chicken. So many buckets of chicken...There was not enough chicken by far. People began to grow impatient. Around the time I finally managed to get on top of the chicken orders, people finally showed up for work and I was free.
Sorry, I was originally only going to write the first part of the comment, but as I wrote it, that day came to mind. I'd forgotten about that.
"McDonald's isn't Amazon" seems like a good enough reason.
Amazon can "just hire a few PhDs/seasoned engineers", plug them into the right places in the company, and expect good things to happen.
McDonald's would need to hire a new team from the leadership down and would be working up-wind in a hot seller's labor market at every level.
> They have 100 (?) menu items, not 1,000,000 products.
McDonald's is a huge company with lots of opportunities to apply ML. Store operation, hiring, supply chain optimization, menu/product design, lots of resource scheduling problems, marketing, etc.
I can easily see McDonald's making back hundreds of millions very quickly just by decreasing the friction and amount of time required to make improvements in these and other areas.
> Is McDonald's the right owner for a high-tech customer intelligence product? It just feels like such a bizarre cultural and core-competency fit.
Automated ordering is becoming a core part of McDonald's customer experience. Automated and augmented decision making is becoming a core part of McDonald's franchise management. Outsourcing those things would mean putting the fate of the company's future in someone else's hands.
Plus, there's huge upside to owning that IP. McDonald's is taking leadership on automating away the sales clerk and decreasing reliance on skilled/knowledgeable franchise owners and managers.
* You want nothing.
* Wrong, I want a burger.
* You want a burger
* What do I want to drink?
* You want a burger
* Wrong, I want a coke
* You want a coke
Two days later.
* What do I want for breakfast?
* You want a burger and a coke.
They need data infrastructure, experts in cleaning data, and actual data scientists, support etc. Its a huge under taking.
Such a team is on the small side for a Fortune 500, but definitely large enough to do real work and more than pay for themselves (especially if you're introducing serious ML into your company for the first time).
That $300mm is an investment in a reoccuring revenue stream as well. There is a very real possibility that the true org cost is zero for MCD.
For companies like these, that operate in 10's of thousands of franchisee-owned locations, the number of products and the combination of configurations is not merely a function of the things you see on the menu. It's all of those products, with their different combination of parts (beef patties, lettuce, etc.), and then franchisee and regional variations (In some countries, you can't tell a franchisee what they can or can't sell, etc.)
Add on top of that customer modifications to the product in their order (extra pickles, no onions, etc.)
Why does this matter? It drives lots of stuff - food costs, inventory, inventory & sales forecasting (how many pickles do I need this week?) new item research, profit margins, etc.
So you take this, multiply it by 10,000, 15,000, or, in McD's case, 36,000 stores across 100+ countries (half both those numbers in my company's case) and you're talking about vast amounts of information across millions of transactions every day, and that is in fact Big Data.
Whether this is a smart move, I'm not sure. I'm of the opinion that if a piece of technology is core to the _operation_ of your business (for a QSR, things like POS systems, backend administrative systems, etc.), it should be in your house. Other things (HR systems, big data systems) should not be. But maybe McD's sees it differently.
don't you mean "and that is in fact Big Mac Data."
This type of data processing across all the McOpCo stores and franchises will be invaluable.
But if you think of them as experts in agriculture, farming, supply chain and logistics, kitchen automation, marketing, global scale business build out... Then I could see how an acquisition like this could be synergistic.
Don't forget that the Trump tax cuts have provided big companies with a windfall. They have way more free cash flow than they did just 2 years ago. So they also are desperate to find new ways to deploy cash.
McDonalds is heavily investing into automation for the ordering process; and as soon as it becomes technically practical and reliable enough, they're going to want to automate even more.
I mean, Amazon also sells books as it used to, and selling stuff is their core business, but they are not limited to that.
Is it right to think that $300M investment + time to turn it into usable product for the restaurants == Loss of many minimum wage (or slightly more) jobs to make back that $300M
My local Wal*Mart now has 25 self checkouts and only 2 people manned registers.
Now it's time for the governments of the world to catch up and introduce a basic income.
McDonalds has been in the real estate game within hospitality larger and longer than anyone.
This move makes a lot of sense and hopefully they move smartly to compete as the industry evolves.
McDonald’s can’t scale ethically raised food to their restaurants at the price points that matter to their customers but they can bring operational and supply chain expertise to startup restaurants.
Maybe this is part of a bigger transition.
Three simultaneous changes are happening related to the availability, personalization, and niche focus of food. I’ve done a deep dive on this (1) if you want to read more.
(1) https://news.ycombinator.com/item?id=19521655