When I pay for food in a restaurant, what I'm paying goes into:
- the cost of food itself
- chef and kitchen staff salaries
- waiter salary/tip
- profit margin
- rent
I'm assuming that the waiter salary/tip will cover delivery costs... so, the cost for the consumer will mostly remain the same.
The cost saving will come from the fact that if the restaurant operates just the kitchen, they don't need excellent location. A part of this saving can be claimed by the company that makes the app.
The rent difference between where you can have a restaurant vs where you can have just a kitchen in the same general area is huge in most places... so I think the business opportunity is real.
Now, the companies are competing to grab market share, and most restaurants still have the same restaurant kitchen, so we haven't seen those changes, but I think it's just a matter of time.
> Now, the companies are competing to grab market share, and most restaurants still have the same restaurant kitchen, so we haven't seen those changes, but I think it's just a matter of time.
I wouldn't be so sure of that. delivery is inherently less profitable for the restaurant. a single employee can serve many more dine-in and especially take-out customers in an hour than they can as a delivery driver. at the same time, you are throwing maintenance and insurance for a vehicle into the equation. you can push a lot of this cost onto the driver, but not all of it. I also doubt that the location savings are that substantial. you don't need to lease the prime real estate that everyone walks by, but you still need to be within a 15-20 minute drive of your customer base. any property within 20 minutes of a dense area tends to be expensive to rent.
there are already a few take-out/delivery only places in my area. the only ones that seem to be doing well have pretty bad food for the price.
> any property within 20 minutes of a dense area tends to be expensive to rent.
As a rule, if one just rents the place on the street behind the one where all the people go, one can cut the rent by 1/3 or 1/2. A kitchen only place can function underground or on a high floor, what takes most of the price out again.
> at the same time, you are throwing maintenance and insurance for a vehicle into the equation.
We're comparing to the customer driving to the restaurant, right? So this was always a cost the customer was paying and it's just moved into the transaction
right, but the fundamental problem is that customers aren't willing to pay the full cost of delivery. delivery places don't just compete with other delivery places on price; they also have to compete with takeout prices. people don't mentally add the cost of gas, vehicle depreciation, and their own time onto takeout prices. plus a pretty common "workflow" is for people to pickup food on their way home from work, which has a negligible cost for them even if they rigorously price it out.
> delivery places don't just compete with other delivery places on price; they also have to compete with takeout prices.
Sure, but right now all of those takeout prices are from restaurants with retail locations. A ghost kitchen that offers delivery could be undercut by a ghost kitchen offering pickup, but in general they'll have a price advantage versus a retail restaurant with a more expensive location and more square footage.
The calculus is especially favorable toward ghost kitchens during pandemic-induced lockdowns, when people aren't otherwise out and about. If I'm driving home from work and can easily stop by a restaurant for takeout, that doesn't cost me much time. But if I'm at home 24/7, then I factor in the time spent going to get food. That makes ghost kitchens with delivery somewhat more attractive.
On one side being home 24/7 gives me a lot more time to cook myself, on the other side going out to buy food finally makes sense, it's one of the few occasions I have now to go out and have a walk in the open.
>> customers aren't willing to pay the full cost of delivery
There is a different market for customers that are willing to pay the cost of delivery, they usually have more mouths to feed, and those mouths are pickier.
If anyone is ever brave enough to actually test this I think they would be surprised, especially during COVID.
Restaurant doesn't deliver? Provide an option to deliver for $20 and I am certain that you will get orders.
If by “cost” you mean car mileage, sure. But in a store pickup situation, there is no additional driver to pay, special packaging for transportation of the food, liability for safe transport of the food, transportation logistics operating costs, etc.
Personally I don't enjoy the dine-in experience that much. It takes an additional hour of time for long meals, mostly just sitting browsing through my phone, or 15-20 minutes for short meals, again browsing through my phone. I don't like the tipping system, feeling pressured to order drinks, or the whole credit card payment ritual. I hope you're right and that we see delivery-focused restaurants in low cost areas with sustainable packaging, so we can enjoy higher-end food at home besides pizza and chinese.
I assume you are dining out alone? Or are you just ignoring your dining companion(s)? I agree dining out alone takes some getting used to, I would normally pony up to the bar and chat with the bartender or other patrons when dining alone.
Yes dining alone, especially on business travel. I just want to eat my entree and go to the hotel. Dining with a group is totally fine as an experience. Dining with a partner is still a bit of a wait if it's someone you're very familiar with. I'd rather have a 1:1 conversation in a quiet private place with a coffee or drink.
America is a weird place. They bring the bill. You put your credit card in the bill. A waiter comes and takes it then brings it back and after all that you have to enter the tip amount.
In Australia they bring the Eftpos machine to your table and I'm pretty sure it's illegal for them to touch your card.
80% of all credit credit card fraud in the US happens at bars and restaurants, and it's because of that ritual. It's why I always pay with cash at restaurants and bars.
That's not true and credit cards are one of the few cases where the fine print does not matter.
That's because by regulation you (the consumer) will not be liable for credit card fraud if they steal the number and other info and use it elsewhere. If they physically steal the card and use it elsewhere your max liability per regulation is $50.
Regardless of fine print these are the max limits of your loss. That's why nobody worries about it.
it's really not such a big deal. if I'm in a hurry, I just give them the card and they come back with a receipt to sign. if there's any issue, I'm confident amex will be happy to process my chargeback.
I think this is more of a cultural thing than anything. if americans are in a rush, we get takeout or go somewhere that doesn't have full table service. I think the "credit card ritual" is usually just a way to show they are not rushing you out the door.
something to understand about america: in general the protections for ordinary folks are pretty bad, but the act of paying for a good/service is sacred. if you (or one of your employees) fuck over a customer, they have ample remedies against you.
What stops the waiter from putting sedatives in your food and then taking all the cash from your wallet when you pass out?
At a certain point you just have to trust that most people aren't trying to scam or rob you. The waiter is too busy putting in other orders and bussing tables to write down all your card details (and I guess social engineer your billing zip code?) I imagine most people who make their living from credit card fraud would rather not spend their days working in a restaurant.
maybe there's something like this buried in the fine print, but I don't think anyone in the US worries about this. I certainly don't. the usual way a transaction goes is you hand the employee your card and they swipe it for you. in any case, it makes no difference whether you wait to get the receipt first for a meal. either way, you are handing them the card and they are swiping it outside your field of view.
credit card issuers heavily favor the customer in disputes. lots of people unknowingly make chargebacks in situations where it's actually not legal to do so. it still gets processed, the customer gets their money back, and the merchant eats a punitive fee.
It's hard to track down. They could just write down your credit card number, name, expiration, and security code. Then use it a month later, at which point you have already been to several restaurants and handed people your card many times. And you don't know if the card was taken in some other way like a online retail hack.
It’s not too difficult for the credit card companies to pin this down. Basic graph algorithms on their datasets of transactions detect this pretty quickly. And these are running constantly trying to detect fraud ASAP. Problematic businesses can lose their credit card privileges too.
Those algorithms can be very good at detecting fraudulent transactions. However, I'm unfamiliar with systems designed to isolate the origin of the fraud. It's not hard trace a stolen card number to a dump of cards sold on the dark web, or the identity of a person using it after the theft, but it is quite hard to figure out how that person came into possession of the card. My subjective impression is that first-party theft + usage of stolen payment credentials is rare--they're usually passed along (sold) to someone else.
In a single isolated incident then yes, it is difficult. However many thrives get greedy and don’t realize it’s fairly easy to identify the point of sale where fraudulent activity originated.
> if there's any issue, I'm confident amex will be happy to process my chargeback.
It's not a huge deal if they run your card for the wrong amount, as you say it can be fixed by AMEX. But if they skim your card and start using it days/weeks later, it's hard to pin down where the fraud started. And you probably have to replace your card, and update the card for vendors that you use auto billpay with.
It's still very new here, and most places just don't do it the way that you all do elsewhere in the world. I think I got my first one last year? And I bank with one of the biggest banks.
Even then, we have just the chip, no PIN.
(There is a PIN for like, ATM transactions, but if I go to Europe, I still have to sign, even if they do the contactless thing. Folks (reasonably) almost never expect it, and have to go find a pen...)
> I'm pretty sure that your agreement with your card company includes that you must not let anyone else touch the card, let alone take it away.
In the US this is the standard way restaurants work, and it is very strange when they do not work this way. At the low end there will sometimes be a cashier who accepts payment.
If I had to guess, I'd say they were talking about (in most US restaurants)
- Ask for check -> waiter brings check -> put card in fold, wait ?? minutes for waiter to return to grab card -> wait ?? minutes for waiter to run card and return it -> sign receipt and leave.
On a busy night, this ritual could last 15-20 minutes.
To be fair, in some countries in Europe (and a small number of restaurants I've been to in the US), it's more like:
- ask for check -> waiter brings hand terminal to table -> swipe card, then leave
To some degree yes, but the market that you are talking about is going to be dominated by repeat diners. It isn't people going out to show off for example - it is people who want want food.
The competition mentality in hospitality businesses is cut-throat and deeply ingrained. If they saw a profit margin, they would attempt to push the delivery service out of the market. If I'm a regular at Toby's Popular Take Away, it seems unlikely that Toby would sit quietly while Uber made money. He would try to muscle in on the distribution too.
It is unlikely that the market share they are fighting for is defensible. A lot of people aren't fighting for it at the moment because there is no profit to be found. The app part might get commoditised or done in a community manner or by some random dude as a side project, it wouldn't be too hard to get local businesses on side with an idea like that. It doesn't need to be a big company.
I don't know if it's everywhere, but here in Germany the profit margins on the food tend to pretty low, and the restaurants try to make up for that by selling drinks pretty expensively.
So for takeout, I just order food, and often just drink tap water. I also assume I'm not the only one doing that, so for the restaurants, a significant part of their profit dries up.
They'll have to adjust to that somehow, most likely by increasing food prices, or by cutting into the extra profit from rent differences that you have laid out. If they do exist, I'm not yet convinced.
In the US ordering (free) water is common for casual restaurants. There is a lot of profit to be made from selling alcoholic drinks, but that’s not in everyone’s business plan.
> I'm assuming that the waiter salary/tip will cover delivery costs...
Unlikely. At least, not a non-exploitative delivery cost. Beyond the employee cost and mileage, there’s the fact that (in the US) delivery drivers are also tipped. They’re also probably more expensive for a company to insure.
If consumers don't drive to get their food, over time their insurance should come down. It should come down more since a restaurant delivery driver can make multiple stops on one optimized route vs a customer just going to the restaurant and back, but once you factor in how much more dangerous scooter delivery is vs car maybe this doesn't hold up.
Why order out to some (probably unsafe) dark kitchen when I have a kitchen in my house?
In this covid times when I don't want to cook I simply go downstairs and order a take out from one of the tens of restaurants around, where I can get the same quality of a dine in at half the price while also supporting local businesses that are suffering.
Food delivery is overrated in my opinion and also unfair to workers.
The only reason I found to order from a delivery company is to skip the waiting line at McDonald's those rare times I feel like having junk food on a Sunday night while watching a movie and don't move from the couch.
But if I want a good pasta I simply order to the guy at the restaurant below my home from the window, don't even need to phone them, and they give me the food in a regular dish, not some lousy plastic wrap that I just return when I have time.
You also need to spend time and money getting to and from the restaurant. If you use Uber both ways it’s already more than
double the cost of getting the food to you.
Having been in the business you way overestimate how much rent actually is. Payment for the kitchen equipment inside is more in many cases. Saving even 1000 a month is not that much. You will spend more than that on your delivery vehicles! (though you can outsource that to the drivers in some cases)
Without having to worry about a store front, you could consolidate kitchen equipment between businesses. Say you use your ovens in the morning to bake bread, and at night to cook meat. Sure, there are limits to this, but there have to be some benefits from centralization.
Another option for cost savings is on the fly demand pricing of meals. Through some process of ingredient availability, customer analytics, registered interest(voting or something), etc, the centralized kitchen could decide to make a particular dish at a greater volume/lower cost which could then be passed on to consumers. i.e. Tuesday is 'spaghetti night' and you can feed your family for 50% less than other options.
I believe this may be how many cafeterias operate. Cafeterias are a good analogy because you have an environment already saving on wait staff and gratuities.
This might seem odd to a typical HN reader who spends $300+ on sushi dinners in SF, but this sort of thing could appeal to many people with less income and without the time to prepare meals themselves. It probably won't work until we can automate the delivery process, but I think at some point the economics will work out in certain markets.
Ghost kitchens are a real thing, but it’s still an open question how much value there is in building a brand relationship woth a customer from a dine-in experience. Cutting out the front-of-house, the ambiance, and the wait staff kinda also cuts out a lot of the personality and connection. Scrolling through a bunch of generic “chicken fettucini alfredo” options on an infinite grid is kinda soulless, and doesn’t give me a sense of what I’m getting for the extra money between the $14 and the $26 versions.
> Say you use your ovens in the morning to bake bread, and at night to cook meat.
There's nothing inherent to the delivery app business that makes this possible. This has always been possible, but it doesn't really happen on any significant scale.
> Without having to worry about a store front, you could consolidate kitchen equipment between businesses.
This is happening right now: see 'ghost kitchens' on delivery apps, with three or four different 'restaurants' that are actually operating out of the same location.
The deck ovens used to bake bread would be challenging to use for other types of food. You could perhaps bake some bread in a combi oven, but it’s going to be much slower.
I'm assuming that the waiter salary/tip will cover delivery costs...
"Will" is doing double duty in that sentence. Maybe with some futuristic delivery tech (Rainbows End-style pneumatic cannons?) that would be possible. Today, delivery costs are covered by VCs and by paying drivers less than minimum wage. (58% of California voters approve!)
> some futuristic delivery tech (Rainbows End-style pneumatic cannons?)
You are closer to the truth than you may realize.
It's a little-known fact that there once was an underground tunnel that delivered piping hot burritos from the SF Bay Area to hungry customers in New Jersey.
It's absolutely ridiculous that Food Delivery apps aren't making gobs of cash right now.
(1) Restaurants are closed for indoor dining (or obviously should be anywhere they aren't). You literally can't sit in a restaurant in good faith, anywhere in the US today.
(2) The Food Delivery app prices are usually 5% to 20% higher than the in-dining counterparts.
(3) In addition to that, the order/delivery fees are much higher than all other competitors. (At least, on my app the 'delivery fee' from GrubHub is 25% higher than the equivalent delivery fee from the local Pizza Hut and Panera Bread, for example)
(4) And because of all of that, the default tip rates are also higher than all other delivery services.
(5) Despite all of the above, everyone is still paying for it. Paid-in-full delivery purchases are at an all-time high, despite all-time high prices for it.
(6) There's a massive record-high unemployment due to people who have lost their jobs from the pandemic! It's cheaper than ever to pick up folks as part-time drivers. And pizza places generally actually employ their drivers, delivery apps lie to the feds and claim their employees as 1099-contractors, cutting their costs even lower.
Sure, DoorDash's "efficiency" claims are all lies. But who cares, they don't need "magically efficient logistics" to make their model work, this model has already worked for every single random pizza place or equivalent from the past 40+ years straight.
These delivery-only-apps have the highest revenue and margins (while having the lowest overhead cost and lowest employee cost) of any food service business I've ever heard of. They've been handed the best possible scenario for their business. How they aren't wildly profitable right now is absolutely due to ridiculous management/financing shenanigans alone, and not because "ordering food for delivery via touch-tone telephone" is some inherently unsustainable business practice.
I would be careful in extrapolating what works for pizza to other types of foods. AFAIK, pizza is very profitable and those margins can subsidize delivery drivers.
Panera Bread offers sandwiches and soups and deli-like food via pizza-like delivery with pizza-like employment (their drivers are real employees, with tips+mileage, health benefits, 401k, PTO, etc)
That kinda works the same, because most of what they're offering is made from bulk, cheap staples that doesn't require that much labor. (Assembly of a sandwich is pretty cheap and you let soups simmer away.)
Most restaurants are not turning out food that is both extremely cheap and fast to make in high volume.
> (6) There's a massive record-high unemployment due to people who have lost their jobs from the pandemic!
That also means that there is less cash around to go to (incredibly expensive) takeout compared to cooking rice & beans at home (where people are now for much longer, without a commute).
And in case you didn't realize, these delivery apps are taking a 20% "marketing fee" out of the food subtotal on every order (in addition to all delivery costs, some of which are paid by the restaurant as well). Just for operating an online menu. How they can be unprofitable with an app that is basically printing money on every order is beyond me.
it's pre-ordered food for a single meal, not a last minute order.
There is a very big single reason the model for last minute order has limited efficiency: how many stops can a delivery driver make before the last delivery is unacceptably late? Basically the suburbs are out of the game, places with prevalent traffic lights are out of the game, that mostly lives bikes/mopeds in the hearts of old world cities.
I was happy to wait an hour and a bit for delivery when dial-a-pizza and the local Chinese restaurant were the only places around that did it.
The food industry is over-optimising for speed. Especially drive-throughs. Every time I've been through one recently the order has come out almost instantly, but made wrong.
If their claimed error rate was true then they would in theory meet part of the six sigma criteria, but the claimed rate was an off the cuff remark in one interview and has never been substantiated.
In India, there are quite a few vertically integrated "cloud kitchen" food delivery startups. I wonder if that's a more sustainable model for food delivery. Most of them (box8.in is a great example) have what appears to be a pretty broad menu, but once you order you can tell they're reusing the same pre-prepared ingredients for most things. It's basically the same as pizza delivery, but for other kinds of food.
> Maybe, as Roy suggested, viable food delivery services look more like Domino’s Pizza or other restaurants that deliver their own food; industrial kitchens that churn out meals only for delivery; or something else that no one can imagine because instead we have food delivery apps that can’t survive on their own.
The problem is that business people have figured out a way to turn the age-old local business-model of a restaurant into a global winner-takes-all game: the one who owns the portal/platform, owns the market.
I think we should really have regulation to break that ridiculous pattern.
I don’t even get delivery directly from a restaurant, preferring to pick up takeaway myself as I can be more assured of having the food have the least possible time sitting around getting cold and mushy. If I were sick is the only remotely common case where I’ll order delivery.
Even there, I’m perfectly capable to call my local restaurant. If someone ends up dominating this field, it’ll be because the local restaurants gave up an enormous advantage they started with.
I don’t see any amount of “optimized route planning” that’s going to get customers food that’s sat around less, tastes better, or costs less which are the only things I can see durably disrupting the “I give you money and you give me food” market.
It's precisely regulation that creates this situation. Creating and operating a delivery platform wouldn't have to be so costly if they didn't need an army of lawyers and lobbyists to simply exist. In a free market, food delivery platforms could be a thin convenience layer charging a small premium instead of the gargantuan 30%+ they are able to command now due to artificial lack of competition.
Amazon was a magical thinking! Google was a magical thinking! iPhone was a magical thinking! This article makes confusing arguments on sustainability and unsustainability.
No, those were different in that there was a path to profitability that didn't involve unicorns and pixie dust.
Uber and Lyft (the progenitors of this model) have been losing money hand over fist for over a decade now, and are both public companies and are still losing money.
Food delivery is incredibly unprofitable (because of the last-mile problem and the fact that it needs to be quick), and Uber have forced otherwise profitable companies to compete with their unprofitable model.
Food delivery from my neighbor (or within my apartment building) is easy. Food delivery from arbitrary locations to arbitrary locations is complex, especially given the product being delivered is time sensitive. Unless the customer is cost blind (i.e. wealthy) and taste ignorant, there is no way to make this work profitably other than inventing a transporter beam.
The complaint is that food delivery is wholly inefficient to the point of unsustainabilty, but then says this is stopping more sustainable options. How so? Thankfully there is nothing other than VC dollars propping this up and so if you can come up with a more efficient approach you will win as there is a proven market it should be easy to come in and out compete.
"The market can remain irrational longer than you can remain solvent." -- J.M. Keynes
Also applies to competitors of Venture funded vortexes like Uber/Lyft/Doordash.
Sure, in the long run your cloud kitchen with guest chefs and nifty social networking features might be a more sustainable model in the long run. But you'll never find out because your effort will be murdered by Uber's marketing spend.
Can somebody explain to me what the problem in the US is? I mean here I choose food, I order it, pay it and it gets delivered in less than 30 minutes (usually). It does not require not exhibit magical thinking to do so. What step does not work in the US? Do people call pizza delivery and before it comes sacrifice some small animals? :-)
Yeah, here in a smaller town in Europe (Czech Republic) you just call to the restaurant (basically 100% pizzerias provide deliveries, others case by case) say what you want & they will tell you an estimate when it will come. Then after a while a guy in a car shows up, you pay him (usually in cash, some accept meal tickets) and you get your food. And the service is not limited to the town but also the villages around it. Some restaurants have their own cars, some cooperate with taxi companies.
A fairly new addition to this are what I would say are "food aggragator services", like: https://www.damejidlo.cz/en/
The provide a common interface for restaurants that cooperate with them in a given area, so you just search the street or location you are in and it will show you the restaurants that provide deliveries in that area & their menus. The main added value is in this case convenience - you can just click what you want and have ti delivered without calling someone. They also have their own cars so they can offer deliveries from restaurants that otherwise don't offer deliveries. Also they make it possible to pay for most deliveries by card.
The downside is of course that not every restaurant that does deliveries is in their system & substantial cut they are taking from the money that goes to the restaurant to cover their services. Which most likely explains why there are restaurants missing in their list.
These delivery/logistics companies are really trying to be marketing companies rather than behind the scenes logistics.
Customers go to the delivery company to order their food rather than to the restaurant. Restaurants pay fees just to be listed.
Delivery company gets a cut of every sale AND they charge the customer a delivery fee AND a "service" fee, which I've found to be roughly 2x the tax rate (which is very high here already). Restaurants a lot of the time make menu items more expensive to offset the cut of the sale that the delivery company takes... which also makes the "service fee" charged to the customer higher. Don't forget, still need to tip the driver, because they are not being paid a living wage.
It is a shit business. Companies trying to be unicorns and don't want to be relegated to being behind the scenes. Even with all the insane fees, they still don't make money.
Make a habit of asking the restaurant how they'd prefer you to place your order.
I don’t know where you’re located, but I live 2 miles from downtown and it takes at least 15 minutes to drive there. So the restaurant would need to see my order, immediately start cooking, and drive immediately to me after it’s done to accomplish the same.
Plenty of other problems, but our urban/suburban sprawl presents challenges.
Well here it is also probably 10min. And yes, the restaurant probably sees the order, immediately starts cooking and immediately drives to me. Why would they delay this process? For the pizza joint one can track the process online and it seems to be frictionless.
I think the "future of food delivery" is larger ghost kitchens, preparing food under different brands to give customer the feeling of choice.
These facilities can share personnel, stock and equipment between different brands. Larger volumes may allow investment in more automation. Facilities can be optimized for quick pickups. The whole food preparation process can be optimized together with deliveries to allow courier to deliver more food on one run.
Any good insights into why food preparation is still stuck in pre-Industrial Revolution technologies? I prepare dinner from my family every night from scratch like I was a medieval artisan. With the economics of industrial production I should be able to purchase pre-made meals that just need to be heated in the oven at one temperature, that are healthy, and cost less than buying components and assembling them. Why hasn’t this happened at scale?
> Any good insights into why food preparation is still stuck in pre-Industrial Revolution technologies? I prepare dinner from my family every night from scratch like I was a medieval artisan.
No you don't. You go to a store and buy processed ingredients that have satisfied numerous legal requirements and inspections to ensure product safety, which are enforced by a vast bureaucracy. The store you bought your ingredients from offered many product options, including food from other countries. In the medieval period you would not be able to afford anything that was not grown further than a few miles from your home unless you were rich.
You prepared those ingredients in your home, which has electricity and running water, and a stove and oven that you can turn on with a knob instead of starting a wood fire. You don't share your home with livestock.
If you are unwilling to prepare your meals with a knife (even though all of the meat has already been cut into "cuts of meat" for you), you can buy food processors, blenders, meat slicers, and other contraptions to make things as easy as possible.
To say that your food preparation is similar to the way people did it before the Industrial Revolution is ridiculous.
The industrial “TV Dinner” has been around since the 1950’s. Pre-made, reheated, and relatively “healthy” in some cases. Have you looked at the frozen aisle of your grocery store?
- the cost of food itself
- chef and kitchen staff salaries
- waiter salary/tip
- profit margin
- rent
I'm assuming that the waiter salary/tip will cover delivery costs... so, the cost for the consumer will mostly remain the same.
The cost saving will come from the fact that if the restaurant operates just the kitchen, they don't need excellent location. A part of this saving can be claimed by the company that makes the app.
The rent difference between where you can have a restaurant vs where you can have just a kitchen in the same general area is huge in most places... so I think the business opportunity is real.
Now, the companies are competing to grab market share, and most restaurants still have the same restaurant kitchen, so we haven't seen those changes, but I think it's just a matter of time.