Now, I mean, would it be a good business that was used by 80% of the population multiple times per week? No. As you point out, it's intractably expensive. But there's nothing wrong with a business that sells a relatively expensive luxury, and it's not sooo expensive to have a meal delivered that only a tiny fraction of the population would ever use it.
Nowadays, I see restaurants charging easily $5-10 above their in-person list price on delivery apps, and some have left entirely now that normalcy has returned and they don’t have to put up with it.
During the pandemic, there was certainly more pressure on the restaurants. After it, the restaurants may get a better deal -- none of that is a particular reason not to comply with the service, it's just one to negotiate more of the costs onto the end user instead of the restaurant.
If they're not, in a world where some mix of users and restaurants get charged the true costs of delivery service, then you're left with a service that gets people to order things for you, which is basically just Postmates, and they're also not doing hot.
It seems clear to me that there is plenty of demand to justify the service. Doordash's revenue was about $5B last year (I believe that's revenue and not GTV, which is probably 5x that). Doordash also only has about 50% of the market. Even if you believe that some of that demand is illusory, brought about by subsidized prices, that still seems like there's a pretty straightforward business there if you can charge a sort of "ordinary business profit."
And, again, there's been delivery food businesses since time immemorial, it would be weird if we thought that there was no kind of business here. Pizza companies used to be more or less the only ones because pizza delivered so well that pizza restaurants could absorb the fixed cost of entire delivery employees, but a food delivery app lets you pay for just part of a delivery-person.
It seems pretty straightforwardly the case to me that the problem in food delivery apps has never been that the fundamental business makes sense, the problem has been that from the very beginning there have been a large number of companies ferociously competing in the space, with nothing to differentiate them, which has squeezed margins to nothing.
Pizza delivery works because there is only one place you need to go to pick up the food that needs to be delivered, and there is a very high volume of food to be delivered. There's no additional traveling cost to flit between restaurants, most of which don't churn out food at high production rates like a pizzeria.
App delivery drivers spend a lot of their time not traveling directly to customers, or waiting around for food to be ready for them to deliver.
---
> Even if you believe that some of that demand is illusory, brought about by subsidized prices, that still seems like there's a pretty straightforward business there if you can charge a sort of "ordinary business profit."
This remains to be seen. MoviePass was never able to successfully both retain customers and raise prices, streaming services are starting to see the same issues. The premium over pickup is often $10-20 already, if it increases anymore that probably starts having a negative exponential effect on market size. (How many people are willing to pay $30? $50? over pickup?)
Restaurant margins are already thin. So they hike the prices to pay UberEats and co. their share.
Also, bigger margin doesn't always mean less effort to cook. It might depending on labor costs and the cost of goods (gravy mix is more expensive than the ingredients, but takes a bit more time to prepare) and you might also wind up needing consistency: Waffle House (a chain restaurant) probably wants its gravy to taste the same in two different cities, and it doesn't matter so much if it takes away a bit of margin.
At that point, though, you're not gigging out delivery people. You're just hiring short-term couriers to move stuff to people from the central kitchens. This is, basically, the model that big city Chinese restaurants and pizza places had anyway before the apps muscled in by swamping them with superior SEO.
There's also a lot of cuisines that simply don't travel well and these Doordash/uber eats business are indifferent. Pizzas travel well, Chinese travels well. Sushi kind of does too assuming you can be quick. But ramen? French fries? Lots of stuff just isn't meant for it.
Amazon won online retail, and changed online retail behavior, by getting rid of shipping fees. The price you see on the product is the price you pay (sans sales tax, but at least in the USA no one is used to seeing tax included).
you actually mean "by hiding shipping fees in an annual subscription model carefully tuned to deal with real world customer ordering behavior".
Non-Prime customers still see shipping fees for everything.
IIUC, 30% is the top end, with rates ranging from 10-30%.
That said, many (most?) restaurants that aren't high-end (which won't be preparing orders for delivery anyway) have gross margins of at most 10-15%.
The cut that DoorDash, GrubHub, etc. take has driven many lower cost restaurants (a number of which had provided their own delivery and been in business for decades) where I live (NYC) out of business over the past 10 years.
The only choice for these folks is to raise prices enough to cover the extra costs, often killing their business (note that GrubHub requires restaurants to charge the same on their site as on their in-house menu, not sure about other services).
The result is fewer independent low-mid priced restaurants. And that's a loss to any community.
Could they charge enough delivery fees to actually build a sustainable business?
How much fees would you need to charge? $4/delivery means having to make a delivery in 15 mins to be able to pay the delivery person minimum wage. Plus Uber’s profit, plus the cost of the car and maintenance expenses + tickets, accidents, etc,
It’s hard to see how a monopoly can also make money, without charging so much that people would rather pick up and restaurants would rather offer their own delivery.