> While the sheer amount of data McDonald’s stored over the years caught me off guard at first, the way it was used by the company’s predictive algorithms was what I found most concerning. The company predicted how often I would visit in the next six weeks (2.16 times), what I would spend on average ($13.49), and what I would spend in total ($29.15).
I always feel a bit disconnected from what other people feel when I read things like this.
Personally, this kind of data doesn’t bother me at all, and is the type of data I don’t mind them having at all, and is the type of analysis I expect them to do. Using the data they have about my direct interactions with them to predict how I interact with them in the future seems fine. As long as they aren’t charging me extra, why should I be upset that they can predict what I will do based on my predictable behavior?
Their job is to provide me a good service that makes me want to keep coming back and giving them my business. If they can use the information from my past interactions with them to provide a better service to me, that seems fine.
My daughter has ARFID, and for about a year McDonald’s chicken nuggets were one of the only safe foods she would eat, so I went quite a few times. I felt much more embarrassed and worried about the workers who saw me there all the time than the app knowing i was there all the time.
What is the actual concern with McDonalds having this data about you? That they will charge you more if they know you will come back anyway?
Data has ruined fast food (among many other businesses).
When I was young it was common for a McDonalds to have 10+ employees working the lunch rush, one for every station and a few floaters cleaning the dining room. Someone took your order right away, and you got your meal in a minute or two.
Now I go and it's 3, sometimes 2 employees. You order on a tablet, and 10+ minutes later an overworked employee sets it on the counter and scurries away, probably after realizing it's not a drive-thru order. The dining room hasn't been cleaned since 6am, the trash cans are full. There's at least one alarm going off constantly.
Back then they were run based on someones intuition of what makes a good customer experience. Now they're run based on the data, and the data says they have enough loyal-to-a-fault customers like the author that chronically understaffing is more profitable than providing a good experience.
The problem with fast feedback loops is that they filter out long term signals. The problem with data driven decision making is that data is numeric, and the act of counting is necessarily an act of approximation, by which we erase the difference between objects or events in order to bucket them into a category so that they can be counted. So the kind of dashboards that modern managers demand, offer nothing but the most obvious, short term "insights", that you might as well eliminate the man in the middle and feed the data directly into a set of simple decision rules.
> The problem with data driven decision making is that data is numeric, and the act of counting is necessarily an act of approximation, by which we erase the difference between objects or events in order to bucket them into a category so that they can be counted.
Yes, this is exactly right. To put it another way: When we quantify something, we abstract away everything about that thing that is not quantifiable. So our thinking is only about a tiny aspect of that thing's existence.
TLDR: Making a decision based on only qualitative data, and therefore ignoring qualitative information and observations, can lead to bad outcomes. Not everything that is important can be easily measured.
> When I was young it was common for a McDonalds to have 10+ employees working the lunch rush, one for every station and a few floaters cleaning the dining room. Someone took your order right away, and you got your meal in a minute or two.
When I was young, they predicted orders and had hot food waiting, you would get your meal as you ordered. Unfortunately, after merciless teasing by other chains, they relented and only assemble to order and cook to order in some cases. Huge increase in lunch latency; ordering via the app can help a little...
It keeps getting worse, too, because everyone in the store is heavily evaluated on the KPI of 'time to serve an order.' But except for drive-thru, the staff control the timer. If you delete the order off the screen, the computer that does all the metrics thinks you've served it.
So of course they delete the orders as soon as humanly possible, relying on the printouts stuck randomly around the kitchen to try to keep track, and now anyone in the management chain who does care about actual service time is completely blinded to how well or poorly they're doing. Even a manager who is out there helping, and can plainly see how slow they are, still knows the metrics are meaningless so it's hard to track progress towards improvement.
Yes, I remember that too. They even had this one promotion back in the day where you could start some timer after you ordered and if you didn't get your meal in 60 seconds you did get something for free (was it the full order? I dont remember).
It's been so obvious for a long long time now, but people still keep believing that smart things make you faster or give you a better life or whatever, but that just doesn't add up. It's just frustrating, often takes longer and the only reason it is there is to grab your data. There is no other reason.
Sure. It's definitely in the "a little." But given how slow the kiosks and apps are to operate, due to all the animations and cross-platform frameworks, getting that order done before you get in the car to go means the order starts when you pull into the parking lot, rather than after you go inside and poke buttons on the kiosk 37 times.
Or in the slim chance that there is a line it's good. But I've rarely seen a number of kiosks insufficient to prevent lines.
The thing I most miss: Being hungry, going into McDonalds and seeing what they had on the shelf ready to eat now. Bic Mac, Quarter Pounder or maybe 2 Hamburgers.
Now? In the queue behind 23 other bespoke orders while they try and service the drive-thru as quickly as possible and nothing stopping them taking the burger that was made for me and adding it to another order.
I was so angry once, I walked out after paying and I just didn't wait for my breakfast muffin because the whole thing like the parent says is that they used to care about efficiency as well as customer experience and now that goes out the window.
The experience now is definitely worse but even back in the early 2000s, owners tracked drive thru service order times and the managers constantly reminded us of this.
I didn't know it at the time, since it was my first job, but they got bonuses based on how fast they could make us work to fulfill those orders during certain times of the day. A backed up drive thru line would always get faster treatment than in store customers. Unless it was a special case, like a bus kid of field trip kids. I assume in those situations the managers could argue for their bonus based on the large amount of sales they got in store.
They obviously didn't share those bonuses with us.
In some ways, it feels like cosmic balance that DT gets priority. Hear me out:
In the DT you are placing yourself at the mercy of every order before you. A delay to the front order at the DT in a peak time delays every car that's entered the queue. Maybe as many as 8 orders. So a delay of one minute there wastes 8 human minutes of life.
At the counter, on the other hand, it's not serial of course, so a delay to your individual order affects only you. So, even by old-fashioned good business sense (rather than modern MBA KPI-optimizing logic), it's reasonable to waste 2 minutes of your time, instead of one minute each of a whole queue at the DT.
(Yes, you can park the delayed car at the DT, but trust me, that's not a sustainable solution to do routinely. In general, getting "behind" during a lunch rush is an extremely tough and frustrating ordeal for the whole crew, no matter what)
This is a silly analogy but it's like integer truncation has been applied everywhere blindly. Staffing model says you need 2.5 employees on average to handle the load? You get 2. Everywhere with everything.
> Now they're run based on the data, and the data says they have enough loyal-to-a-fault customers like the author that chronically understaffing is more profitable than providing a good experience.
If they are indeed saying understaffing doesn't matter because their current customers don't care, their loyalty numbers will only go up because of that sampling bias. Relatively upscale fast casual places (e.g. Chipotle) will fill replace McDonalds for the customers they lose. The circle of service industry life.
Different stores in different chains have a very different attitude. There is a Taco Bell in Ithaca which has a staff that has some esprit de corps and then another where the dining room is disgusting.
what you are describing is basically company ownership or high level managers of the store are just keeping kicking out legs of the table and waiting to see if it will still stand up
because you and others keep going back to the store no matter how bad it gets, they keep testing what other cuts they can do
eventually it will be just one employee sitting in the corner watching several humanoid bots doing all the work and just required to press the reset button when they glitch
maybe even a single human remote employee doing that for 100+ stores
vandalism will get insane and then corporations will just get congress to pass felony laws to slow that
I've seen the same changes. The most noticeable is, now that they are no longer required to be visible at the front counter in case someone comes in to order, because of kiosks, the workers usually specifically hide anywhere else, to minimize the chances of being disturbed by customers with their endless requests ("I've been waiting for 15 minutes for a large fry...")
Don't underestimate the equally intense downward pressure on staffing that minimum-wage hikes have provided (for instance, in California). I won't speculate if it's been "worth it" since it'll devolve into an off-topic political debate, but basic economics does predict that since labor costs twice what it did 10 years ago, they'll do everything to minimize how much labor they buy. Couple that with what you said, basically, using "Moneyball" techniques to finesse everything with all these optimizations, we get the operations we see now in all retail establishments.
And even if there are states that haven't done the minimum wage changes, the markets that have are so big. The national businesses that have adapted to it have proven out their barebones-labor model there - why wouldn't they roll out most of the changes nationwide?
I think if someone launched a fast-food place or a drugstore in 1995 that operates with the type of skeleton crew they use today, it would fail overnight due to people having a choice. But today, since essentially 100% of retail operates this same way, they don't experience any pressure to do better.
> Don't underestimate the equally intense downward pressure on staffing that minimum-wage hikes have provided (for instance, in California).
In-N-Out allows us to discard this notion. They've maintained full staffing while simultaneously paying staff above-market rates (even in California). They only charge 30 cents more for a basic cheeseburger, and the burger is much higher quality, so even that is hard to attribute to wages.
In-N-Out is, IIRC, completely privately owned by the founder's family. Every location is directly controlled. Their customers are the people buying burgers.
McDonalds, of course, is 5% corp-owned flagships and 95% franchise "opportunities". They don't have customers, they have tenant farmers who work the land leased to them.
In-N-Out is a low-margin, high-volume business, kind of like how Walmart is.
They literally can't cut staff because they only make money due to the incredibly high number of meals they serve per hour. So they have to serve as efficiently as possible, which requires that fully-staffed kitchen and counter, or there wouldn't be enough time in the mealtime hours to do that volume otherwise. It's just as impossible for them to cut any staff as an airline could remove pilots from the cockpit, in that the whole operation would cease to function.
McDonald's and the rest are never going to be In-N-Out though because it would take years of operating at steep losses in the hopes people would notice how fast and tasty they'd become. I can't think of a mature company that's ever pulled off going from "boring mainstream place" to "cult favorite" like that.
Is it not common knowledge that this is the entire purpose of loyalty programs? As far as I know, if you don’t give them your details and just swipe a card, they are not doing facial recognition and PCI stops them from linking your purchases too heavily.
I’m not really surprised by anything here but the fact that McDonalds disclosed all their predictions?
The bigger question here is whether it is good for large corporations to be run with such brutal profit motives, running on extremely precise data to uncover how to best make a bigger profit. It doesn’t appear good to me, but it is how the entire US economy operates. The lift from the per-customer loyalty timeseries and per-customer upsells is probably modest compared to the rest of the data driven work to understand how to sell more “product”.
> Privacy experts I spoke with said this level of detail may feel invasive, but it's fairly standard for how large companies in the US run their loyalty programs.
Something can be both invasive and standard. This practice is both.
What level of detail is invasive here though? How often and which items a customer has ordered and their predictions on future customer behaviour based on that data?
I understand privacy concerns in many cases, especially when it comes to big tech and online privacy, but this to me looks like a complete nothingburger.
I see a lot of "don't worry about privacy invasion X, because privacy invasion Y is unavoidable." when talking about this stuff. People are being trained to accept "normal" violations
In general, I agree with you and I'm a privacy advocate. In this particular case, though, it seems like we're railing against the idea that a retailer with repeat customers keeps track of what those customers are buying most frequently and how much they spend. It would be very surprising if retailers didn't retain that info, since it's data owned by the company, generated by the core business itself, and useful for forecasting. I'm having trouble understanding the "invasive" part.
I grew up in New York state, but my dad used to shop from a company called Cable Car Clothiers, based out of San Francisco, and he'd order everything by mail or over the phone. He probably shopped there for 20 years or more before I moved to San Francisco and I literally walked by that very shop on my way to work. He knew that my new job took me by their storefront and so he asked me to go in and pick something up for him. I had a short conversation with the shop owner who had a retail system that showed every order my dad had ever placed going all the way back to the late 80s. There is a bunch of derivative data that is available when you have a log like this, like the total amount ever spent, the most frequently purchased item, date of the most recent purchase, date of the oldest purchase, etc.
Is it the scale of McDonalds that's causing concern, or is it something else?
>“McDonald’s secret sauce is really commercial surveillance,”
Journo is search of a narrative that isn't there. McDonalds could stop tracking customers tomorrow and people would still be eating the same food there in 50 years.
The secret sauce is has more to do with actual sauce than anything claimed here.
"When I spoke with Cranor, she was floored by just how often I went to McDonald’s: “This guy eats a lot of McDonald's.” My boss, who was also stunned by my processed-food choices, says I now have to stop eating there. Despite my love of crispy Diet Cokes and greasy french fries, I agree."
I wonder how many of us would stop going not just to McDonald’s, but all the other places if we knew just how much they ALL have in our files. They're not tracking us to be helpful, they're tracking us to increase their profits.
At my first start up, I unfortunately relied on fast food daily. One night I pulled into a Wendy’s and the drive thru operator opened with “Hey Mike. Working late again, eh? The usual [my usual order]”. I don't recall having a friendly relationship with the operator or really talking to any one in particular, so this freaked me out and also made me feel a bit of shame. I stopped going after that. Granted, this was just a friendly worker but your question reminded me of it.
Definitely prompted me to remove the Taco Bell app. I don’t need the occasional free bean burrito that badly.
One of the people working at the coffee place my wife and I stop at after grocery shopping remembered my name and order and I thought it was pretty nice to be greeted by name. They aren't there anymore and I miss it.
It does sound nice for a local shop. And I don’t discredit the worker for being nice in my case. I was however disturbed by the fact I was shamefully slamming that much fast food at 1 AM every day.
From what was listed in the article it seemed like it was all data generated by the chain. If it included a lot of stuff from outside sources I would probably be put off, but this seems fairly benign.
I suppose they could be sharing that information with third parties and that could be bad.
It’s nefarious by definition, because the actual full statement is
> they’re tracking us to increase their profits, everything else be damned. Including your interests, your rights, the law, morality, and long term side effects.
Yeah I agree, from what was in the article it looked like stuff anyone would want to know about their customers. How often they visit, top items ordered, and average spending per order.
It could even be helpful, if they know what you usually order they could present that as a "one-click" option rather than you having to navigate their incredibly tedious standard kiosk UI to order what you want.
I'm sure they already are. Can you imagine how much health insurance companies would luv to get their hands on such data? How much they would be willing to pay?
Crisp (and refreshing) is a perfectly fine way to describe a drink. “Crispy” doesn’t quite read the same but I can see a way it becomes the shortened usage of whole phrase.
Similarly this reminds of Target's surveillance system, which is loved and adopted by police and federal agencies (https://unicornriot.ninja/21st-century-jim-crow-a-series/). Honestly it's one of the more technically advanced things about them or any store.
What an awful article. Just spent like 10 minutes digging through a pile of ideologically loaded buzzwords and implied accusations of racism trying to find anything concrete the authors are accusing Target of actually doing. Closest thing I can find is a vague allusion to Target donating money to a city to fund security cameras?
I've seen a lot of stories about how Target's loss prevention staff will track thefts over multiple visits and don't get the police involved until the loss passes some threshold.
I've always wondered how they are able to get convictions from a theft months ago with no evidence that you actually possessed the item after you left the store? Is it really good enough to just have video that looks like you probably stole something and an inventory system that they claim shows lost inventory?
> I've always wondered how they are able to get convictions from a theft months ago with no evidence that you actually possessed the item after you left the store?
If you steal a car and dump it in the sea, but there's camera footage of you stealing the car, is the camera footage not enough evidence even if they don't find the car?
I suppose also if they have video over months showing a pattern of a person appearing to steal it might be enough to get past any claims that you ditched the item before leaving many times over a period of months.
At first I thought "building a 515-page dossier" was needlessly hyperbolic for what amounts to a printed PDF of typical CRM values from a database. It's not like a McDonald's detective meticulously crafted it.
However, companies do have an absurd amount of data on us, and the fact that it's automatically collected makes it more troubling, not less. But it's normalized.
>the fact that it's automatically collected makes it more troubling, not less
I don't think so. You can inflate raw data arbitrarily. They could have laid out on a page the decimal byte values of a one minute video clip of you waiting in line or something. Something redacted by a human would be much more dense in terms of relevant information, and much more digested. If it's just a data dump then at least it's information that not necessarily anyone knows, in the sense of residing in someone's brain.
Information collection is also not necessarily malicious. You can never know what you might need in the future. I was kind of on board with Recall, if the data could live in storage in my control and can I could turn it on and off at will.
> Information collection is also not necessarily malicious
Anyone collecting information has an obligation to think about how it could be used in the future - whether by government or by corporation.
McDonald's has also very visibly and successfully switched to gamifying their order selection with machines and dark patterns to drive people to higher priced items, so it's safe for anyone there to assume any data collected will be used to harm customers in such ways.
In my area, the majority of McDonald's customers use the drive thru, and the dining room isn't very busy. So for the price of a coffee I can get a couple hours of work done from a table near a power outlet and wifi.
My only complaint is their firewall blocks the tailscale control plane.
Of all the places I could spend some time to work, a greasy, run-down, dirt-cheap fast food "restaurant" really is the last option… I’d pay more voluntarily to not spend a second there longer than necessary.
Traffic monitoring for a fast food chain is dubious at best even in the unencrypted days, but basically useless since the advent of HTTPS and TLS everywhere.
I connect to work over a traditional VPN. Is there some way for me to send that tunnel through Tailscale? I want to do that because our VPN whitelists IPs.
Well... that's what it means to opt into convenience and benefits. That loyalty program that benefits the customer over the years of course also benefits the company, and not just by making said customer more "loyal".
> I requested a copy of my data from McDonald’s loyalty program and received an extensive, personalized report that algorithmically predicts my next purchase.
I mean, that is exactly what i would think a loyalty program would contain.
What is the point of predictive analytics for them? Surely they don't care about what one specific customer will do, only in aggregate? Since ML models can only learn from past aggregates, wouldn't traditional statistics on past transactions make more sense?
Yeah, the discussion is not differentiating between data that's collected, meaning it's not their data (and so they'd have to ask you for it), and data that is generated by the business itself, which is their data. Businesses knowing what frequent customers buy most often and how much they spend goes back hundreds of years, I'd expect.
"She says a more beneficial approach for consumers would be highly detailed disclosures during the sign-up process that lay out what’s going to be stored in your 'permanent record' and how that data will be used to make specific inferences about you."
No. You should never have to try to figure out what companies are collecting or how they are using it. They should not ever be allowed to aggregate data at the individual level. And if aggregate data ever becomes fine-grained enough to fingerprint individuals, then regulation should force coarser graining. (This would mean that better machine learning systems would result in companies being able to collect less data.) We should have forced audits of every consumer company.
Some of the collection is creepy, and I’ve largely stopped using mobile apps and loyalty programs myself. However, some of the concern seems very misplaced.
“Oh no McDonald’s runs a Customer Lifetime Value algorithm, a thing in use for decades and which has minimal privacy implications.”
“GASP the Golden Arches knows I like Diet Coke and has applied a marketing segmentation to me!”
Congrats, you’ve discovered they use concepts taught in Marketing 101?
Yeah, it’s 500+ pages, where is the surprising stuff? I assume the author of the piece would have chosen to list out the most egregious/shocking data out of the document, right?
I don’t think any of this is non-surprising in the “we’ve been conditioned to think this is normal” kind of way. This is really just a basic transaction history with extremely basic non-AI analytics applied.
The length of the document sounds scary but it also seems like it’s essentially a transaction list.
I would have thought the article might get into more details like location tracking like being able to predict where that person would be at a certain time, or maybe details inferred from data like “this person buys McFlurries when they’re on their period.”
Something scarier would be figuring our friend and family relationships, where relatives and friends live, etc.
I’m sure that could be done with data like this, and it’s critical that companies like McDonald’s keep that data secure and delete it when they don’t need it. We should have better laws in that regard.
At some point McDonald’s doesn’t really care about all that extra creepy stuff. Anything beyond figuring out your buying preferences is not useful information to them. The biggest concern would be that they would be willing to sell that data to someone else who is interested.
That’s where I’m at. I really don’t care if an individual business collects usage data like this. I have a problem when that data is used outside of that context.
Try to sell me more burgers and fries? Fine, but don’t tell my doctor, or palantir, or the government etc etc. I don’t believe that’s a huge ask as a consumer in the Information Age.
This is a very fair take. McDonalds is entitled to the data about their customers' transactions with the business because that's the business. If they turned around and started selling it to health insurance, I'd have a problem with that.
the vast majority of people have not taken Marketing 101, and it'd be strange to expect them to have. I know you're not literally talking about a Marketing 101 course, but the point is the same
I’m not talking about a literal Marketing 101 course, but I’m also not entirely facetious about this all being taught in a first or second Marketing class (having attended those).
I agree we should not expect everyone to have taken a marketing class - but I sure expect the author of the piece to have demonstrated some curiosity about some of the terms. And we know they did, because the author decoded them into plainer language in their table. So why present these elements as an abnormal thing for McDonald’s to compute about them, rather than something an undergrad might be expected to compute as part of a project?
Yeah I walked into this with my pitchfork ready but isn't this stuff pretty... reasonable?
McDonalds is just a restaurant, they predict what menu items you like and how likely you will come back, is that really a problem? What is the concern here? I don't use their app but presumably for those that do, isn't this exactly the type of data that would simply influence whether the app offers a coupon or something? Or how well a franchise is doing? Maybe also what to keep on the menu or how much to stock?
What exactly is the bar here I wonder? McDonalds has security cameras, it also has menu kiosks with cameras (I think?). The alternate reality evil tech-like McDonalds could have theoretically used facial recognition to categorize all footage of any given customer, then used this footage to estimate their friends, spouse, BMI, age, gender, ethnic group, attractivness, disability, etc. They could then sell it to third party data brokers for extra cash.
That would certainly be the kind of dystopian surveillance to get me swinging my pitchfork around, but this? Having assumed the worst I'm now walking away realizing McDonalds might not be so bad after all.
The concern is when the data breaks out of the silos.
It should not shock anyone that Amazon has an entire history of everything that you've ever bought. I would likely not surprise anyone here that they have a log of every item you've clicked on, or even moused over. The beauty of modern systems is that storing all of that data is very cheap.
The dark side is if the fellow queried McDonalds and found not just his McDonalds trends, but his shopping habits from, say, Target. Where McDs and Target perhaps had a data sharing agreement to correlate customers.
Or seeing a copy of his credit report in there buried underneath the stack of Diet Coke and Quarter Pounder receipts.
That's where the concern shows up. Not the raw data, but the cross cutting concerns as the data is potentially more centralized. Perhaps into a company that you don't even know is collecting info on you. Some corp that McDs is sharing with. You request your dossier from McDs because you know they're collecting your data. But you don't request it from Company X, because you don't know they're doing so.
Then there's the issue of handing the data over to the authorities without so much as a warrant -- different problem.
What people think of these programs is often linked to the perception of the product being sold. If a cigarette company did this, people would be pissed. McDonalds is kind of an edge case - everyone knows it's not great for you, but it's not as harmful as cigarettes or drinking so you get mixed takes.
The classic case here is Target inferring that someone was pregnant before her family did[1] - so yeah, this is nothing new.
Where it gets "you really should care about this" are things like your insurance carrier using this sort of data (which is for sale to all comers) to, say, increase your premiums under the guise of "your 2 Big Mac a week habit makes you a bigger risk".
https://archive.md/gkqRQ
> While the sheer amount of data McDonald’s stored over the years caught me off guard at first, the way it was used by the company’s predictive algorithms was what I found most concerning. The company predicted how often I would visit in the next six weeks (2.16 times), what I would spend on average ($13.49), and what I would spend in total ($29.15).
I always feel a bit disconnected from what other people feel when I read things like this.
Personally, this kind of data doesn’t bother me at all, and is the type of data I don’t mind them having at all, and is the type of analysis I expect them to do. Using the data they have about my direct interactions with them to predict how I interact with them in the future seems fine. As long as they aren’t charging me extra, why should I be upset that they can predict what I will do based on my predictable behavior?
Their job is to provide me a good service that makes me want to keep coming back and giving them my business. If they can use the information from my past interactions with them to provide a better service to me, that seems fine.
My daughter has ARFID, and for about a year McDonald’s chicken nuggets were one of the only safe foods she would eat, so I went quite a few times. I felt much more embarrassed and worried about the workers who saw me there all the time than the app knowing i was there all the time.
What is the actual concern with McDonalds having this data about you? That they will charge you more if they know you will come back anyway?
Data has ruined fast food (among many other businesses).
When I was young it was common for a McDonalds to have 10+ employees working the lunch rush, one for every station and a few floaters cleaning the dining room. Someone took your order right away, and you got your meal in a minute or two.
Now I go and it's 3, sometimes 2 employees. You order on a tablet, and 10+ minutes later an overworked employee sets it on the counter and scurries away, probably after realizing it's not a drive-thru order. The dining room hasn't been cleaned since 6am, the trash cans are full. There's at least one alarm going off constantly.
Back then they were run based on someones intuition of what makes a good customer experience. Now they're run based on the data, and the data says they have enough loyal-to-a-fault customers like the author that chronically understaffing is more profitable than providing a good experience.
The problem with fast feedback loops is that they filter out long term signals. The problem with data driven decision making is that data is numeric, and the act of counting is necessarily an act of approximation, by which we erase the difference between objects or events in order to bucket them into a category so that they can be counted. So the kind of dashboards that modern managers demand, offer nothing but the most obvious, short term "insights", that you might as well eliminate the man in the middle and feed the data directly into a set of simple decision rules.
> The problem with data driven decision making is that data is numeric, and the act of counting is necessarily an act of approximation, by which we erase the difference between objects or events in order to bucket them into a category so that they can be counted.
Yes, this is exactly right. To put it another way: When we quantify something, we abstract away everything about that thing that is not quantifiable. So our thinking is only about a tiny aspect of that thing's existence.
Everyone quotes Goodhart's law, but I think the McNamara fallacy is even more important nowadays. It should be read aloud to every CEO every day.
https://en.wikipedia.org/wiki/McNamara_fallacy
TLDR: Making a decision based on only qualitative data, and therefore ignoring qualitative information and observations, can lead to bad outcomes. Not everything that is important can be easily measured.
> Making a decision based on only qualitative data
I suppose you meant quantitative and it's just a typo
> When I was young it was common for a McDonalds to have 10+ employees working the lunch rush, one for every station and a few floaters cleaning the dining room. Someone took your order right away, and you got your meal in a minute or two.
When I was young, they predicted orders and had hot food waiting, you would get your meal as you ordered. Unfortunately, after merciless teasing by other chains, they relented and only assemble to order and cook to order in some cases. Huge increase in lunch latency; ordering via the app can help a little...
It keeps getting worse, too, because everyone in the store is heavily evaluated on the KPI of 'time to serve an order.' But except for drive-thru, the staff control the timer. If you delete the order off the screen, the computer that does all the metrics thinks you've served it.
So of course they delete the orders as soon as humanly possible, relying on the printouts stuck randomly around the kitchen to try to keep track, and now anyone in the management chain who does care about actual service time is completely blinded to how well or poorly they're doing. Even a manager who is out there helping, and can plainly see how slow they are, still knows the metrics are meaningless so it's hard to track progress towards improvement.
Yes, I remember that too. They even had this one promotion back in the day where you could start some timer after you ordered and if you didn't get your meal in 60 seconds you did get something for free (was it the full order? I dont remember).
It's been so obvious for a long long time now, but people still keep believing that smart things make you faster or give you a better life or whatever, but that just doesn't add up. It's just frustrating, often takes longer and the only reason it is there is to grab your data. There is no other reason.
Edit: Found it. You did get a coke for free. Damn, 2004. https://www.burgerschachteln.de/2004/Sonstiges/index.htm
TIL there's burgerschachteln.de, because of course there is a person collecting burger packaging on the Internet.
The machines were right; truly civilization peaked in 1999.
> ordering via the app can help a little...
It can? I used it once a few years ago and they wouldn't start making the food until I was in the parking lot.
Sure. It's definitely in the "a little." But given how slow the kiosks and apps are to operate, due to all the animations and cross-platform frameworks, getting that order done before you get in the car to go means the order starts when you pull into the parking lot, rather than after you go inside and poke buttons on the kiosk 37 times.
Or in the slim chance that there is a line it's good. But I've rarely seen a number of kiosks insufficient to prevent lines.
The thing I most miss: Being hungry, going into McDonalds and seeing what they had on the shelf ready to eat now. Bic Mac, Quarter Pounder or maybe 2 Hamburgers.
Now? In the queue behind 23 other bespoke orders while they try and service the drive-thru as quickly as possible and nothing stopping them taking the burger that was made for me and adding it to another order.
I was so angry once, I walked out after paying and I just didn't wait for my breakfast muffin because the whole thing like the parent says is that they used to care about efficiency as well as customer experience and now that goes out the window.
The experience now is definitely worse but even back in the early 2000s, owners tracked drive thru service order times and the managers constantly reminded us of this.
I didn't know it at the time, since it was my first job, but they got bonuses based on how fast they could make us work to fulfill those orders during certain times of the day. A backed up drive thru line would always get faster treatment than in store customers. Unless it was a special case, like a bus kid of field trip kids. I assume in those situations the managers could argue for their bonus based on the large amount of sales they got in store.
They obviously didn't share those bonuses with us.
In some ways, it feels like cosmic balance that DT gets priority. Hear me out:
In the DT you are placing yourself at the mercy of every order before you. A delay to the front order at the DT in a peak time delays every car that's entered the queue. Maybe as many as 8 orders. So a delay of one minute there wastes 8 human minutes of life.
At the counter, on the other hand, it's not serial of course, so a delay to your individual order affects only you. So, even by old-fashioned good business sense (rather than modern MBA KPI-optimizing logic), it's reasonable to waste 2 minutes of your time, instead of one minute each of a whole queue at the DT.
(Yes, you can park the delayed car at the DT, but trust me, that's not a sustainable solution to do routinely. In general, getting "behind" during a lunch rush is an extremely tough and frustrating ordeal for the whole crew, no matter what)
My read on this, beyond data and MBA's, is that all of the slack has been pulled out of the economy.
This is a silly analogy but it's like integer truncation has been applied everywhere blindly. Staffing model says you need 2.5 employees on average to handle the load? You get 2. Everywhere with everything.
> Now they're run based on the data, and the data says they have enough loyal-to-a-fault customers like the author that chronically understaffing is more profitable than providing a good experience.
If they are indeed saying understaffing doesn't matter because their current customers don't care, their loyalty numbers will only go up because of that sampling bias. Relatively upscale fast casual places (e.g. Chipotle) will fill replace McDonalds for the customers they lose. The circle of service industry life.
Different stores in different chains have a very different attitude. There is a Taco Bell in Ithaca which has a staff that has some esprit de corps and then another where the dining room is disgusting.
what you are describing is basically company ownership or high level managers of the store are just keeping kicking out legs of the table and waiting to see if it will still stand up
because you and others keep going back to the store no matter how bad it gets, they keep testing what other cuts they can do
eventually it will be just one employee sitting in the corner watching several humanoid bots doing all the work and just required to press the reset button when they glitch
maybe even a single human remote employee doing that for 100+ stores
vandalism will get insane and then corporations will just get congress to pass felony laws to slow that
$20 burgers will also slow that
I miss pizza
I've seen the same changes. The most noticeable is, now that they are no longer required to be visible at the front counter in case someone comes in to order, because of kiosks, the workers usually specifically hide anywhere else, to minimize the chances of being disturbed by customers with their endless requests ("I've been waiting for 15 minutes for a large fry...")
Don't underestimate the equally intense downward pressure on staffing that minimum-wage hikes have provided (for instance, in California). I won't speculate if it's been "worth it" since it'll devolve into an off-topic political debate, but basic economics does predict that since labor costs twice what it did 10 years ago, they'll do everything to minimize how much labor they buy. Couple that with what you said, basically, using "Moneyball" techniques to finesse everything with all these optimizations, we get the operations we see now in all retail establishments.
And even if there are states that haven't done the minimum wage changes, the markets that have are so big. The national businesses that have adapted to it have proven out their barebones-labor model there - why wouldn't they roll out most of the changes nationwide?
I think if someone launched a fast-food place or a drugstore in 1995 that operates with the type of skeleton crew they use today, it would fail overnight due to people having a choice. But today, since essentially 100% of retail operates this same way, they don't experience any pressure to do better.
> Don't underestimate the equally intense downward pressure on staffing that minimum-wage hikes have provided (for instance, in California).
In-N-Out allows us to discard this notion. They've maintained full staffing while simultaneously paying staff above-market rates (even in California). They only charge 30 cents more for a basic cheeseburger, and the burger is much higher quality, so even that is hard to attribute to wages.
In-N-Out is, IIRC, completely privately owned by the founder's family. Every location is directly controlled. Their customers are the people buying burgers.
McDonalds, of course, is 5% corp-owned flagships and 95% franchise "opportunities". They don't have customers, they have tenant farmers who work the land leased to them.
In-N-Out is a low-margin, high-volume business, kind of like how Walmart is.
They literally can't cut staff because they only make money due to the incredibly high number of meals they serve per hour. So they have to serve as efficiently as possible, which requires that fully-staffed kitchen and counter, or there wouldn't be enough time in the mealtime hours to do that volume otherwise. It's just as impossible for them to cut any staff as an airline could remove pilots from the cockpit, in that the whole operation would cease to function.
McDonald's and the rest are never going to be In-N-Out though because it would take years of operating at steep losses in the hopes people would notice how fast and tasty they'd become. I can't think of a mature company that's ever pulled off going from "boring mainstream place" to "cult favorite" like that.
Is it not common knowledge that this is the entire purpose of loyalty programs? As far as I know, if you don’t give them your details and just swipe a card, they are not doing facial recognition and PCI stops them from linking your purchases too heavily.
I’m not really surprised by anything here but the fact that McDonalds disclosed all their predictions?
The bigger question here is whether it is good for large corporations to be run with such brutal profit motives, running on extremely precise data to uncover how to best make a bigger profit. It doesn’t appear good to me, but it is how the entire US economy operates. The lift from the per-customer loyalty timeseries and per-customer upsells is probably modest compared to the rest of the data driven work to understand how to sell more “product”.
What is the word “but” doing in this sentence?
> Privacy experts I spoke with said this level of detail may feel invasive, but it's fairly standard for how large companies in the US run their loyalty programs.
Something can be both invasive and standard. This practice is both.
What level of detail is invasive here though? How often and which items a customer has ordered and their predictions on future customer behaviour based on that data?
I understand privacy concerns in many cases, especially when it comes to big tech and online privacy, but this to me looks like a complete nothingburger.
Subtly normalizing the invasive data collection practice.
Or maybe not so subtly.
I see a lot of "don't worry about privacy invasion X, because privacy invasion Y is unavoidable." when talking about this stuff. People are being trained to accept "normal" violations
In general, I agree with you and I'm a privacy advocate. In this particular case, though, it seems like we're railing against the idea that a retailer with repeat customers keeps track of what those customers are buying most frequently and how much they spend. It would be very surprising if retailers didn't retain that info, since it's data owned by the company, generated by the core business itself, and useful for forecasting. I'm having trouble understanding the "invasive" part.
I grew up in New York state, but my dad used to shop from a company called Cable Car Clothiers, based out of San Francisco, and he'd order everything by mail or over the phone. He probably shopped there for 20 years or more before I moved to San Francisco and I literally walked by that very shop on my way to work. He knew that my new job took me by their storefront and so he asked me to go in and pick something up for him. I had a short conversation with the shop owner who had a retail system that showed every order my dad had ever placed going all the way back to the late 80s. There is a bunch of derivative data that is available when you have a log like this, like the total amount ever spent, the most frequently purchased item, date of the most recent purchase, date of the oldest purchase, etc.
Is it the scale of McDonalds that's causing concern, or is it something else?
It means you don't need to worry, just trust the mega corporation to do what's best for you with this data!
>“McDonald’s secret sauce is really commercial surveillance,”
Journo is search of a narrative that isn't there. McDonalds could stop tracking customers tomorrow and people would still be eating the same food there in 50 years.
The secret sauce is has more to do with actual sauce than anything claimed here.
At my first start up, I unfortunately relied on fast food daily. One night I pulled into a Wendy’s and the drive thru operator opened with “Hey Mike. Working late again, eh? The usual [my usual order]”. I don't recall having a friendly relationship with the operator or really talking to any one in particular, so this freaked me out and also made me feel a bit of shame. I stopped going after that. Granted, this was just a friendly worker but your question reminded me of it.
Definitely prompted me to remove the Taco Bell app. I don’t need the occasional free bean burrito that badly.
One of the people working at the coffee place my wife and I stop at after grocery shopping remembered my name and order and I thought it was pretty nice to be greeted by name. They aren't there anymore and I miss it.
It does sound nice for a local shop. And I don’t discredit the worker for being nice in my case. I was however disturbed by the fact I was shamefully slamming that much fast food at 1 AM every day.
User name checks out!
> they're tracking us to increase their profits
That's not really all that nefarious is it?
From what was listed in the article it seemed like it was all data generated by the chain. If it included a lot of stuff from outside sources I would probably be put off, but this seems fairly benign.
I suppose they could be sharing that information with third parties and that could be bad.
It’s nefarious by definition, because the actual full statement is
> they’re tracking us to increase their profits, everything else be damned. Including your interests, your rights, the law, morality, and long term side effects.
The last bit is unspoken but ALWAYS there.
Yeah I agree, from what was in the article it looked like stuff anyone would want to know about their customers. How often they visit, top items ordered, and average spending per order.
It could even be helpful, if they know what you usually order they could present that as a "one-click" option rather than you having to navigate their incredibly tedious standard kiosk UI to order what you want.
it'll become nefarious when they start selling it to health insurance companies.
When?
I'm sure they already are. Can you imagine how much health insurance companies would luv to get their hands on such data? How much they would be willing to pay?
It's creepy
I used to eat McD’s once a week as my chest meal. Fries, coke, the lot. I loved it.
Then my Mum got cancer, and my brother just said “why would you put that shit in your body?”
That was seven years ago and I’ve never had fast food or sofa since. I can’t imagine how sick it would make me feel.
i think they mean "crispy" fries, and "greasy" drinks?
Crisp (and refreshing) is a perfectly fine way to describe a drink. “Crispy” doesn’t quite read the same but I can see a way it becomes the shortened usage of whole phrase.
a crisp drink, sure; in a similar vein to carrot juice though, calling a carrot 'juicy' is cromulent.... but....
Similarly this reminds of Target's surveillance system, which is loved and adopted by police and federal agencies (https://unicornriot.ninja/21st-century-jim-crow-a-series/). Honestly it's one of the more technically advanced things about them or any store.
What an awful article. Just spent like 10 minutes digging through a pile of ideologically loaded buzzwords and implied accusations of racism trying to find anything concrete the authors are accusing Target of actually doing. Closest thing I can find is a vague allusion to Target donating money to a city to fund security cameras?
I've seen a lot of stories about how Target's loss prevention staff will track thefts over multiple visits and don't get the police involved until the loss passes some threshold.
I've always wondered how they are able to get convictions from a theft months ago with no evidence that you actually possessed the item after you left the store? Is it really good enough to just have video that looks like you probably stole something and an inventory system that they claim shows lost inventory?
> I've always wondered how they are able to get convictions from a theft months ago with no evidence that you actually possessed the item after you left the store?
If you steal a car and dump it in the sea, but there's camera footage of you stealing the car, is the camera footage not enough evidence even if they don't find the car?
Do you think the people getting caught are going to trial or just taking plea deals?
I suppose also if they have video over months showing a pattern of a person appearing to steal it might be enough to get past any claims that you ditched the item before leaving many times over a period of months.
At first I thought "building a 515-page dossier" was needlessly hyperbolic for what amounts to a printed PDF of typical CRM values from a database. It's not like a McDonald's detective meticulously crafted it.
However, companies do have an absurd amount of data on us, and the fact that it's automatically collected makes it more troubling, not less. But it's normalized.
>the fact that it's automatically collected makes it more troubling, not less
I don't think so. You can inflate raw data arbitrarily. They could have laid out on a page the decimal byte values of a one minute video clip of you waiting in line or something. Something redacted by a human would be much more dense in terms of relevant information, and much more digested. If it's just a data dump then at least it's information that not necessarily anyone knows, in the sense of residing in someone's brain.
Information collection is also not necessarily malicious. You can never know what you might need in the future. I was kind of on board with Recall, if the data could live in storage in my control and can I could turn it on and off at will.
> Information collection is also not necessarily malicious
Anyone collecting information has an obligation to think about how it could be used in the future - whether by government or by corporation.
McDonald's has also very visibly and successfully switched to gamifying their order selection with machines and dark patterns to drive people to higher priced items, so it's safe for anyone there to assume any data collected will be used to harm customers in such ways.
In my area, the majority of McDonald's customers use the drive thru, and the dining room isn't very busy. So for the price of a coffee I can get a couple hours of work done from a table near a power outlet and wifi.
My only complaint is their firewall blocks the tailscale control plane.
Of all the places I could spend some time to work, a greasy, run-down, dirt-cheap fast food "restaurant" really is the last option… I’d pay more voluntarily to not spend a second there longer than necessary.
What about a well maintained, bright and cheerful fast food "restaurant" with really fast wifi?
I wonder if they monitor your internet traffic.
Traffic monitoring for a fast food chain is dubious at best even in the unencrypted days, but basically useless since the advent of HTTPS and TLS everywhere.
> My only complaint is their firewall blocks the tailscale control plane.
Brother, did you give your brain to AI already? There exists technology
If you have a suggestion, you should give it.
If you initiate the connection on your cell phone hotspot and then connect to WiFi, it’ll continue working.
I connect to work over a traditional VPN. Is there some way for me to send that tunnel through Tailscale? I want to do that because our VPN whitelists IPs.
Irony: The first thing shown to me when opening that Wired article was an info box asking me to agreeing to data collection.
Well... that's what it means to opt into convenience and benefits. That loyalty program that benefits the customer over the years of course also benefits the company, and not just by making said customer more "loyal".
> I requested a copy of my data from McDonald’s loyalty program and received an extensive, personalized report that algorithmically predicts my next purchase.
I mean, that is exactly what i would think a loyalty program would contain.
What is the point of predictive analytics for them? Surely they don't care about what one specific customer will do, only in aggregate? Since ML models can only learn from past aggregates, wouldn't traditional statistics on past transactions make more sense?
They can see if marketing changes your behavior.
"We predict he won't get fries until Monday. Send a push notification."
How many pages would be required to store all the information that a shopkeeper knows about one of their regulars?
Yeah, the discussion is not differentiating between data that's collected, meaning it's not their data (and so they'd have to ask you for it), and data that is generated by the business itself, which is their data. Businesses knowing what frequent customers buy most often and how much they spend goes back hundreds of years, I'd expect.
> “McDonald’s secret sauce is really commercial surveillance,” says Jeff Chester
Great line. I do really love their fries though.
"She says a more beneficial approach for consumers would be highly detailed disclosures during the sign-up process that lay out what’s going to be stored in your 'permanent record' and how that data will be used to make specific inferences about you."
No. You should never have to try to figure out what companies are collecting or how they are using it. They should not ever be allowed to aggregate data at the individual level. And if aggregate data ever becomes fine-grained enough to fingerprint individuals, then regulation should force coarser graining. (This would mean that better machine learning systems would result in companies being able to collect less data.) We should have forced audits of every consumer company.
There is an 18MB gif playing on this page. What the hell is wrong with Wired?
"I was concerned about being surveilled so I also uploaded all my data to an LLM company"
Could solve this, and in turn rid the world of a huge amount of algorithmic slop, by taxing companies according to how much personal data they hold.
Some of the collection is creepy, and I’ve largely stopped using mobile apps and loyalty programs myself. However, some of the concern seems very misplaced.
“Oh no McDonald’s runs a Customer Lifetime Value algorithm, a thing in use for decades and which has minimal privacy implications.”
“GASP the Golden Arches knows I like Diet Coke and has applied a marketing segmentation to me!”
Congrats, you’ve discovered they use concepts taught in Marketing 101?
Yeah, it’s 500+ pages, where is the surprising stuff? I assume the author of the piece would have chosen to list out the most egregious/shocking data out of the document, right?
I don’t think any of this is non-surprising in the “we’ve been conditioned to think this is normal” kind of way. This is really just a basic transaction history with extremely basic non-AI analytics applied.
The length of the document sounds scary but it also seems like it’s essentially a transaction list.
I would have thought the article might get into more details like location tracking like being able to predict where that person would be at a certain time, or maybe details inferred from data like “this person buys McFlurries when they’re on their period.”
Something scarier would be figuring our friend and family relationships, where relatives and friends live, etc.
I’m sure that could be done with data like this, and it’s critical that companies like McDonald’s keep that data secure and delete it when they don’t need it. We should have better laws in that regard.
At some point McDonald’s doesn’t really care about all that extra creepy stuff. Anything beyond figuring out your buying preferences is not useful information to them. The biggest concern would be that they would be willing to sell that data to someone else who is interested.
That’s where I’m at. I really don’t care if an individual business collects usage data like this. I have a problem when that data is used outside of that context.
Try to sell me more burgers and fries? Fine, but don’t tell my doctor, or palantir, or the government etc etc. I don’t believe that’s a huge ask as a consumer in the Information Age.
This is a very fair take. McDonalds is entitled to the data about their customers' transactions with the business because that's the business. If they turned around and started selling it to health insurance, I'd have a problem with that.
Information asymmetry is ok for me but not for thee.
Yeah I kept reading the article and thinking ... "This seems reasonable." "This seems fine."
It also wasn't clear how much of the data they have is because he's using the app. If it is, it seems much/most of the data was in his control.
Do they think McDonalds (or whichever company) is going to spend a lot of money building an app and a rewards program and not expect an ROI on it?
the vast majority of people have not taken Marketing 101, and it'd be strange to expect them to have. I know you're not literally talking about a Marketing 101 course, but the point is the same
I’m not talking about a literal Marketing 101 course, but I’m also not entirely facetious about this all being taught in a first or second Marketing class (having attended those).
I agree we should not expect everyone to have taken a marketing class - but I sure expect the author of the piece to have demonstrated some curiosity about some of the terms. And we know they did, because the author decoded them into plainer language in their table. So why present these elements as an abnormal thing for McDonald’s to compute about them, rather than something an undergrad might be expected to compute as part of a project?
Yeah I walked into this with my pitchfork ready but isn't this stuff pretty... reasonable?
McDonalds is just a restaurant, they predict what menu items you like and how likely you will come back, is that really a problem? What is the concern here? I don't use their app but presumably for those that do, isn't this exactly the type of data that would simply influence whether the app offers a coupon or something? Or how well a franchise is doing? Maybe also what to keep on the menu or how much to stock?
What exactly is the bar here I wonder? McDonalds has security cameras, it also has menu kiosks with cameras (I think?). The alternate reality evil tech-like McDonalds could have theoretically used facial recognition to categorize all footage of any given customer, then used this footage to estimate their friends, spouse, BMI, age, gender, ethnic group, attractivness, disability, etc. They could then sell it to third party data brokers for extra cash.
That would certainly be the kind of dystopian surveillance to get me swinging my pitchfork around, but this? Having assumed the worst I'm now walking away realizing McDonalds might not be so bad after all.
If the concern is misplaced, where should it be placed?
The concern is when the data breaks out of the silos.
It should not shock anyone that Amazon has an entire history of everything that you've ever bought. I would likely not surprise anyone here that they have a log of every item you've clicked on, or even moused over. The beauty of modern systems is that storing all of that data is very cheap.
The dark side is if the fellow queried McDonalds and found not just his McDonalds trends, but his shopping habits from, say, Target. Where McDs and Target perhaps had a data sharing agreement to correlate customers.
Or seeing a copy of his credit report in there buried underneath the stack of Diet Coke and Quarter Pounder receipts.
That's where the concern shows up. Not the raw data, but the cross cutting concerns as the data is potentially more centralized. Perhaps into a company that you don't even know is collecting info on you. Some corp that McDs is sharing with. You request your dossier from McDs because you know they're collecting your data. But you don't request it from Company X, because you don't know they're doing so.
Then there's the issue of handing the data over to the authorities without so much as a warrant -- different problem.
focusing on the weeds in your lawn would be more productive
What people think of these programs is often linked to the perception of the product being sold. If a cigarette company did this, people would be pissed. McDonalds is kind of an edge case - everyone knows it's not great for you, but it's not as harmful as cigarettes or drinking so you get mixed takes.
The classic case here is Target inferring that someone was pregnant before her family did[1] - so yeah, this is nothing new.
[1] https://www.forbes.com/sites/kashmirhill/2012/02/16/how-targ...
Where it gets "you really should care about this" are things like your insurance carrier using this sort of data (which is for sale to all comers) to, say, increase your premiums under the guise of "your 2 Big Mac a week habit makes you a bigger risk".
Insurance is about accurately pricing risk, not subsidizing covert risky behaviors. Why shouldn't they consider your diet?
Altria already does this to some extent. See their Digital Trade Program.