Well you're adding two parties to the transaction. If you go to McDonald's you just have to pay McDonald's. If you use a delivery app you also have to pay the delivery person and the app.
Well you’re really adding one party. They raise the price a little on the items, maybe a service fee. Fees. Because the food can’t deliver itself. (This is for the best.) Underhanded approach, but not insane reasoning.
Unfortunately… after baking in some income for the extra party, the delivery service…Well, said party wants to shift the burden of compensating empl… contractors to you. Through a tip! That’s fun. You can show your… appreciation. For them doing their job. Super motivating for drivers. Especially when everyone tips well.
Oh…
They brought the third party. Didn’t have to.
Could it be two parties? Sure. Pizza places, all you have to pay are the pizza place. And the driver they also didn’t wish to compensate fairly. But not all restaurants deliver.
Big Mac is $8 with tax. Let’s say Uber takes $4. If I texted you right now and said I’d pay you $8 to stop what you’re doing, drive to McDonalds, then drive to a stranger’s home 2 miles away, would you do it? Not a lot of people would, there are nice jobs in America.
They do. Just good business sense. Why your order might be cold or incomplete. Well, one reason.
Hopefully the drivers are fairly compensated. (They aren’t, hence the driver tip.)
Delivery of food from arbitrary restaurants so you do not to leave the house, by companies underpaying drivers while raising prices silently, adding service fees, and trying to push the burden of compensating the driver on you via tips.
First world problems require first world solutions.
Big Mac is $8 with tax. If I texted you right now and said I’d pay you $8 to stop what you’re doing, drive to grocery store to get buns, then get back to your kitchen and make the burger, would you do it? Not a lot of people would, there are nice jobs in America.
Chefs (by which we really mean line cooks) don't drive to get ingredients and then go back to your kitchen. It takes about 4 minutes to make a big mac, and you're likely going to be cooking more than one big mac at a time. They also don't work on demand so this premise:
> Big Mac is $8 with tax. If I texted you right now and said I’d pay you $8 to stop what you’re doing,
is nonsense. They get paid a fixed rate regardless of how many orders they fill.
The original premise was not that nobody is willing to do demand work for marginal wages, it's that there's enough competition in the labor market that wages need to be competitive, on a per hour basis, for things to work out. And that these one off jobs tend to need to be pretty pricey to make up for the variance in work availability vs. a steady hourly pay.
Most pizza places that do delivery have inhouse full-time pizza delivery drivers. The margins on pizza is so large, delivery can be done at cost (+ tips) and the pizza store can still come out on top.
Pizza travels well, is still good (maybe better) after it cools a bit (but still warm), and has enough volume demand for delivery.
Burgers are ok a bit cold but still warm, but fries are not.
A pizza place with a driver on staff can do route planning and coordination so maybe the pizzas for a group of three deliveries near each other come out of the oven around the same time, hopefully right as the driver gets back from their previous run.
That's harder to do with McDonald's. And if there's much of a wait, the fries show up cold and it's a bad experience.
Funny question. I always wonder, why are delivery apps so cheap. Here in Prague, i can get 6 USD pizza delivered for free, which seems absolutely insane.
I find them almost weirdly cheap; I would not deliver anything (with some exceptions) for 10x the amount (the delivery people get after paying everything).
Edit; checked here; 1 Big Mac large menu is 11.63 euros including delivery. 7.65 euros for takeaway of the same menu. Distance similar to yours.
If I were to guess: there is likely some margin to make delivery cheaper. However, the volume won't be there to make up for it.
The volume won't be there to make up for it because all non-free delivery is too expensive for a good many people. Quite likely there is a price point at/below which delivery will start to appeal to larger segment; perhaps at that price, almost everyone will opt for delivery instead of going there. That price is likely below the costs of providing delivery.
It's probably better to just focus on and capture the market segment that forks up for delivery (because it's worth it to them) rather than chasing an elusive volume increase at lower margins.
If they could knock a dollar off delivery but triple the volume by doing that, they probably would; and likely they've already done that sort of price optimization.
Another thing to consider is order size. If you had a big group of people and had 10 of these meals delivered, maybe the economics would be better?
Answers here are missing the point. McDonald's is 2.2 miles away from you. But the delivery driver is not at McDonalds. They need to go to McDonald's, wait for your order to be finished, probably pick up other orders, and then go deliver it to you. Then they need to wait for another order. If they're coming in hot its probably no big deal. If they actually need to wait then their driver income goes down per hour. Which means Uber needs to pay them more per order to make this worthwhile to engage with.
If McDonalds had people on site to deliver it would be cheaper. But it still has to be pricey enough that they can afford to pay the delivery guy, per hour, enough to make it worthwhile.
The delivery apps here have people at hotspots. I don't know if they're actually told to wait there by the app or if it's their own personal initiative. But McDonald's and malls often have riders camping there.
It might be deliberate as we see some of them riding bicycles for short range deliveries.
Uber has been very poor at optimizing logistics here, which is why they lost. I talked with drivers who were pissed at having to drive 30 mins for a 15 min delivery, and not be told where it was. But it could be a mix of paying drivers too little, resulting in too few people and long waiting times, resulting in people buying from the competition.
> Uber has been very poor at optimizing logistics here, which is why they lost. I talked with drivers who were pissed at having to drive 30 mins for a 15 min delivery, and not be told where it was. But it could be a mix of paying drivers too little, resulting in too few people and long waiting times, resulting in people buying from the competition.
No, it's the logistical problem not making sense. The reason they have to drive so far for a 15 minute delivery is because there's nobody closer. The reason they don't tell the drivers is because they wouldn't want to do some of these drives if they had all of the information. Uber is likely making pretty good guesses at this problem, but it's a very hard problem that frequently won't have a feasible solution that is profitable.
You have a driver at A. You get an order for B. Do you send the driver to B or wait for an order to A? Hmm crap there's nobody else nearby. Shit we just got an order for B that's going in the opposite direction. Should we bundle them up?
The key problem is the ratio of drivers to demand. If the ratio is very high, the optimization is easy but the drivers spend a lot of time not doing orders and so they waste time and earn a shitty wage. Wait time is high.
If the ratio is very low, the optimization is very difficult and requires a lot of guesswork. And by guesswork that's queuing theory and maybe some time series forecasting combined with real time optimization algorithms. Wait times for drivers are lower but delivery times get a lot worse and the labor is more expensive.
The only time it "works" is when everyone is sweating. There's just enough drivers to meet demand and get a satisfying wage. And all of that depends on the price for sending drivers on mostly continuous routes is palatable to the customer, which it likely isn't because frankly that sucks and will always have a lot of variance for the drivers.
Put another way. If a driver wants to earn X per hour, if they spend 10 minutes waiting per hour, they need to be paid 20% more than X per hour for the hours they actually work (6/5). If they spend 20 minutes waiting per hour, they need to be paid 50% more than X per hour for the hours they actually work (6/4). The uncertainty and difficulty of the optimization problem mean that they're going to have these breaks. So even if your mental model says "I would pay someone $X per hour to do deliveries" you need to put on a pretty hefty percentage for the burden of the gig model.
And then of course uber is pretty bloated and unprofitable.
Want to ask the same question but opposite: Why are deliveries so cheap in China? Sometimes it's even cheaper than in the stores. Sometimes when I walk past a store I would go home and order that delivered to get it cheaper. Not at China rn so can't really order one and check the receipt.
Pizza places are an example of what an efficient (low-cost) implementation might look like. They often bring in too many drivers for the dinner rush and let people go as demand tapers off, they'll cook pizzas in the right order so that a driver taking a double can have both items come off the line at once, and they have loading/serving stations optimized for getting the driver out the door faster. The net result is that drivers are making plenty of money since they're constantly busy, while there are usually enough of them to keep people from getting cold food.
Even in those circumstances, the fully loaded cost of the driver to the customer in some no-name suburbs was around $18/hr, split between on average no more than 3 customers. So on a $5 pizza you'd pay $11.50 for the pizza, tax, and delivery costs.
Contrast that model a bit with delivery apps:
(1) Doubles are rare, so the driver earns less per hour.
(2) There's still a back-and-forth between housing and food, but extra latency was tossed in since you have to wait after your delivery to figure out which place you're going to be grabbing food from. I.e., drivers make less per hour.
(3) Enough restaurants aren't fully optimized for fast deliveries that the drivers easily tack on an extra 5 minutes waiting for their food, resulting again in fewer deliveries per hour and less money.
(4) We can keep going on about why drivers are getting fewer deliveries per hour, but the consequence is that drivers need more compensation per order to be willing to do that over another job. Easily 50% more. Your $6 delivery cost is now $9.
(5) Combine that with the fact that they know the tip ahead of time. There are basically no consequences for skipping an order you don't want to take, so the only people getting hot food are on the high end of the tip scale, which is already higher than the pizza place average. Conservatively, bump this up to $12 per delivery, but when I tried DoorDash I was making $36+/hr for at most 2.2 deliveries per hour. You do the math.
(6) On top of all that, there's a middleman in the transaction with their own fixed costs they need to cover. The pizza place is happy just to sell more pizzas per hour by having a delivery option and charges a fee large enough to roughly offset the cost to them of providing drivers (included in that initial $6 estimate I provided). Uber Eats can't increase their profits _just_ by increasing orders per hour since they didn't have a cut of any of the initial transaction. They have to charge their own fees on top of everything else you're already paying.
I'm actually a little surprised the total bill was only $20. Was it cold?
Well, that depends a lot on where you are. Here in Bangkok, Thailand, that would cost $5.86 with standard delivery, or $6.28 with priority delivery. Basically comes down to the fact that labor costs are much lower here (minimum wage is $10 a day); delivery is done via scooter, rather than car (lower fuel costs); population density is high (restaurants are close to customers).
40 comments
[ 2.4 ms ] story [ 83.4 ms ] threadUnfortunately… after baking in some income for the extra party, the delivery service…Well, said party wants to shift the burden of compensating empl… contractors to you. Through a tip! That’s fun. You can show your… appreciation. For them doing their job. Super motivating for drivers. Especially when everyone tips well.
Oh…
They brought the third party. Didn’t have to.
Could it be two parties? Sure. Pizza places, all you have to pay are the pizza place. And the driver they also didn’t wish to compensate fairly. But not all restaurants deliver.
Hopefully the drivers are fairly compensated. (They aren’t, hence the driver tip.)
Delivery of food from arbitrary restaurants so you do not to leave the house, by companies underpaying drivers while raising prices silently, adding service fees, and trying to push the burden of compensating the driver on you via tips.
First world problems require first world solutions.
> Big Mac is $8 with tax. If I texted you right now and said I’d pay you $8 to stop what you’re doing,
is nonsense. They get paid a fixed rate regardless of how many orders they fill.
The original premise was not that nobody is willing to do demand work for marginal wages, it's that there's enough competition in the labor market that wages need to be competitive, on a per hour basis, for things to work out. And that these one off jobs tend to need to be pretty pricey to make up for the variance in work availability vs. a steady hourly pay.
Burgers are ok a bit cold but still warm, but fries are not.
A pizza place with a driver on staff can do route planning and coordination so maybe the pizzas for a group of three deliveries near each other come out of the oven around the same time, hopefully right as the driver gets back from their previous run.
That's harder to do with McDonald's. And if there's much of a wait, the fries show up cold and it's a bad experience.
+1.95 menu price on DD. +$3.99 delivery fee +$3.53 Fees(15%) & tax +$3.00 Tip
pretty sure the drivers only getting like $6.5
Not sure why they need 10%-20% from the restaurant and charge us $(3.99-6.99)+15%
How much did it literally cost to have it delivered? (Assuming it was not $20 on the dot)
How much extra beyond what you would have paid in person do you belief it should cost and why?
Edit; checked here; 1 Big Mac large menu is 11.63 euros including delivery. 7.65 euros for takeaway of the same menu. Distance similar to yours.
2. How much do you think it should cost, exactly? Do you mean to imply that the cost structure of such services should be much lower?
The volume won't be there to make up for it because all non-free delivery is too expensive for a good many people. Quite likely there is a price point at/below which delivery will start to appeal to larger segment; perhaps at that price, almost everyone will opt for delivery instead of going there. That price is likely below the costs of providing delivery.
It's probably better to just focus on and capture the market segment that forks up for delivery (because it's worth it to them) rather than chasing an elusive volume increase at lower margins.
If they could knock a dollar off delivery but triple the volume by doing that, they probably would; and likely they've already done that sort of price optimization.
Another thing to consider is order size. If you had a big group of people and had 10 of these meals delivered, maybe the economics would be better?
If McDonalds had people on site to deliver it would be cheaper. But it still has to be pricey enough that they can afford to pay the delivery guy, per hour, enough to make it worthwhile.
It might be deliberate as we see some of them riding bicycles for short range deliveries.
Uber has been very poor at optimizing logistics here, which is why they lost. I talked with drivers who were pissed at having to drive 30 mins for a 15 min delivery, and not be told where it was. But it could be a mix of paying drivers too little, resulting in too few people and long waiting times, resulting in people buying from the competition.
No, it's the logistical problem not making sense. The reason they have to drive so far for a 15 minute delivery is because there's nobody closer. The reason they don't tell the drivers is because they wouldn't want to do some of these drives if they had all of the information. Uber is likely making pretty good guesses at this problem, but it's a very hard problem that frequently won't have a feasible solution that is profitable.
You have a driver at A. You get an order for B. Do you send the driver to B or wait for an order to A? Hmm crap there's nobody else nearby. Shit we just got an order for B that's going in the opposite direction. Should we bundle them up?
The key problem is the ratio of drivers to demand. If the ratio is very high, the optimization is easy but the drivers spend a lot of time not doing orders and so they waste time and earn a shitty wage. Wait time is high.
If the ratio is very low, the optimization is very difficult and requires a lot of guesswork. And by guesswork that's queuing theory and maybe some time series forecasting combined with real time optimization algorithms. Wait times for drivers are lower but delivery times get a lot worse and the labor is more expensive.
The only time it "works" is when everyone is sweating. There's just enough drivers to meet demand and get a satisfying wage. And all of that depends on the price for sending drivers on mostly continuous routes is palatable to the customer, which it likely isn't because frankly that sucks and will always have a lot of variance for the drivers.
Put another way. If a driver wants to earn X per hour, if they spend 10 minutes waiting per hour, they need to be paid 20% more than X per hour for the hours they actually work (6/5). If they spend 20 minutes waiting per hour, they need to be paid 50% more than X per hour for the hours they actually work (6/4). The uncertainty and difficulty of the optimization problem mean that they're going to have these breaks. So even if your mental model says "I would pay someone $X per hour to do deliveries" you need to put on a pretty hefty percentage for the burden of the gig model.
And then of course uber is pretty bloated and unprofitable.
Even in those circumstances, the fully loaded cost of the driver to the customer in some no-name suburbs was around $18/hr, split between on average no more than 3 customers. So on a $5 pizza you'd pay $11.50 for the pizza, tax, and delivery costs.
Contrast that model a bit with delivery apps:
(1) Doubles are rare, so the driver earns less per hour.
(2) There's still a back-and-forth between housing and food, but extra latency was tossed in since you have to wait after your delivery to figure out which place you're going to be grabbing food from. I.e., drivers make less per hour.
(3) Enough restaurants aren't fully optimized for fast deliveries that the drivers easily tack on an extra 5 minutes waiting for their food, resulting again in fewer deliveries per hour and less money.
(4) We can keep going on about why drivers are getting fewer deliveries per hour, but the consequence is that drivers need more compensation per order to be willing to do that over another job. Easily 50% more. Your $6 delivery cost is now $9.
(5) Combine that with the fact that they know the tip ahead of time. There are basically no consequences for skipping an order you don't want to take, so the only people getting hot food are on the high end of the tip scale, which is already higher than the pizza place average. Conservatively, bump this up to $12 per delivery, but when I tried DoorDash I was making $36+/hr for at most 2.2 deliveries per hour. You do the math.
(6) On top of all that, there's a middleman in the transaction with their own fixed costs they need to cover. The pizza place is happy just to sell more pizzas per hour by having a delivery option and charges a fee large enough to roughly offset the cost to them of providing drivers (included in that initial $6 estimate I provided). Uber Eats can't increase their profits _just_ by increasing orders per hour since they didn't have a cut of any of the initial transaction. They have to charge their own fees on top of everything else you're already paying.
I'm actually a little surprised the total bill was only $20. Was it cold?