4 ms·

Thought exercise: I’ve often thought that some sort of “CDN for food” would be necessary to make food delivery competitive with groceries and home cooked meals.

Thought exercise: I’ve often thought that some sort of “CDN for food” would be necessary to make food delivery competitive with groceries and home cooked meals. How it would work is: 1. You tell the service that within a certain time slot, you’d like your order of [insert order here] to be delivered. 2. All your neighbours do the same 3. At the time slot, a vehicle (or robot) deposits the food in some kind of storage locker for you and several of your neighbours to pick up. This would substantially drive down the per-order delivery cost. A dense city where everyone lives in apartments, such as New York, would be an ideal place for this idea, since orders for everyone in your building could be dropped off at one point

tryplane 6y ago
Exactly.. I work in food tech and batching is a MASSIVE focus for both delivery products and ghost kitchens (colocating many restaurants) because of the kind of efficiencies you can squeeze out of it.
crazygringo 6y ago
It sounds great in theory. The problem is that people want 30 different types of food (Chinese, Italian, Ethiopian), they don't want it cold, and while NYC is dense it's not dense enough. Even in a building with 100 apartments (which is much larger than the norm), at most maybe 2 apartments will want the same type of food in the same half-hour slot. It's not going to be much different from what a delivery driver already does, which is take out 3 orders from the same restaurant to 3 different buildings at the same time. And remember that's just the delivery aspect. The costs of food prep are massive too.
- petra 6y ago
  What if you had 30 different restaurants at the same place ?
  - spideymans 6y ago
    Exactly what the OP article is about, but yes, that would be perfect for this type of application. Cloud kitchens, with dozens of restaurants, delivering food to a single drop point, serving dozens of residents at once, would enormously drive down delivery costs. Likely to well under a dollar per order. I think the biggest stumbling block with this idea is that enormous amount of customers that would be needed to make this operation efficient. You’d need to convince hundreds of people within a neighbourhood or apartment block to give up cooking, and rely on the delivery service
devchix 6y ago
Have you ever come across a restaurant that had "Ordinary" in its name and wondered why? "Ordinary" was a set-meal served at a certain time in a pub, for a fixed price. The logical progression for what you proposed is a neighborhood canteen on wheels. The appeal of takeaways/deliveries is anything you want anytime you want it. The two don't converge.
- smoe 6y ago
  They can converge. In Colombia, where instead of office cafeterias/canteens you have tiny independent set-meal/fixed price restaurants everywhere (often multiples per block). Even before Covid you could get subscriptions to have a lunch delivered to your office every day. If more people work from home in the future it could make sense as an alternative to the office canteen, not the traditional anything anytime takeaways.

tryplane 6y ago

Exactly.. I work in food tech and batching is a MASSIVE focus for both delivery products and ghost kitchens (colocating many restaurants) because of the kind of efficiencies you can squeeze out of it.

crazygringo 6y ago

It sounds great in theory. The problem is that people want 30 different types of food (Chinese, Italian, Ethiopian), they don't want it cold, and while NYC is dense it's not dense enough. Even in a building with 100 apartments (which is much larger than the norm), at most maybe 2 apartments will want the same type of food in the same half-hour slot. It's not going to be much different from what a delivery driver already does, which is take out 3 orders from the same restaurant to 3 different buildings at the same time. And remember that's just the delivery aspect. The costs of food prep are massive too.

petra 6y ago
What if you had 30 different restaurants at the same place ?
- spideymans 6y ago
  Exactly what the OP article is about, but yes, that would be perfect for this type of application. Cloud kitchens, with dozens of restaurants, delivering food to a single drop point, serving dozens of residents at once, would enormously drive down delivery costs. Likely to well under a dollar per order. I think the biggest stumbling block with this idea is that enormous amount of customers that would be needed to make this operation efficient. You’d need to convince hundreds of people within a neighbourhood or apartment block to give up cooking, and rely on the delivery service

devchix 6y ago

Have you ever come across a restaurant that had "Ordinary" in its name and wondered why? "Ordinary" was a set-meal served at a certain time in a pub, for a fixed price. The logical progression for what you proposed is a neighborhood canteen on wheels. The appeal of takeaways/deliveries is anything you want anytime you want it. The two don't converge.

smoe 6y ago
They can converge. In Colombia, where instead of office cafeterias/canteens you have tiny independent set-meal/fixed price restaurants everywhere (often multiples per block). Even before Covid you could get subscriptions to have a lunch delivered to your office every day. If more people work from home in the future it could make sense as an alternative to the office canteen, not the traditional anything anytime takeaways.