3 ms·

I was visiting Palo Alto one afternoon last year and as we were grabbing dinner one thing I noticed (from the tshirts) were all the "dashers" waiting for take a

I was visiting Palo Alto one afternoon last year and as we were grabbing dinner one thing I noticed (from the tshirts) were all the "dashers" waiting for take away just like we were. Considering all the time we had to wait I could not help but think that, surely, it would have been more efficient to schedule other deliveries in the mean time and just come pick up the food when it was ready? As a logistics company you want to minimize the time your couriers sit around... The kitchen already knows roughly how long it's going to take anyway, and with enough data computers could do that automatically too, I guess all delivery services work the same but I never use them, do other incumbents do these things more efficiently nowadays?

danieltillett 11y ago
This is the exact thinking that got all these companies and investors into this niche. Unfortunately, the restaurant business is too messy to automate (at least right now) and too prone to falling into profitless competition. There is a great business in this area waiting to be unleashed, but I have no idea how anyone can make a profit.
jimkri 11y ago
The "dashers" could also keep track of the wait times for each restaurant, and then work accordingly to the wait times. They could boost their delivery's, but it definitely would take several weeks to get the right amount of knowledge about each restaurant.
mathattack 11y ago
The listed wait times on the App are 60-70 minutes. As a former New Yorker (used to bike delivery of sub-30 minutes) it seems crazy that they could have a viable business. It's even more surprising given that they don't have food warmers, so the food arrives cold. This tells me that they're not really in the Food Business - they're in the delivery business, and learning how to do it with food.
potatolicious 11y ago
The trick with DoorDash is that many (most?) of the restaurants in the system aren't voluntarily participating with DoorDash. A smart system that summoned delivery people when the food was ready/close to ready would require integration on the restaurant's part (see: Seamless/GrubHub/Caviar), but DoorDash seems to be going in a different direction. This has been a point of controversy - DoorDash offers delivery for restaurants that do not participate in delivery by simply sending someone there, ordering, and bringing you the food. This is terrible for the delivery people obviously, and DoorDash has also been caught marking up the prices on these menu items without informing the customer. Overall, highly doubtful about the sustainability of this business model. Or the ethics. Side note: Postmates also does similar - some of the restaurants in their system are not official partners, and so the delivery person gets to make the order and wait around. This is a large part of why I stopped using them - they're getting paid peanuts as it is and now they're sitting on their ass for 20+ minutes to bring someone $20 of food and foregoing much needed tips.
bhahn 11y ago
(disclaimer: I work for DoorDash but have minimal exposure to our scheduling platform) > it would have been more efficient to schedule other deliveries in the mean time and just come pick up the food when it was ready? There are a lot of factors at play so it's hard to say what exactly happened in this instance, but some considerations: * The wait time at a restaurant is probably not enough time to fit comfortably another delivery. Deliveries take a bit of time. I don't have the exact #'s, but you can imagine 5-20 minutes driving to restaurant (+ parking), 0-5 minutes waiting at a restaurant, 5-20 minutes driving from restaurant to customer, which ends up being around 10-45 minutes with high variance. * We want food to be as fresh as possible for the customer, so on average we may prefer a dasher to wait a little rather than arrive the food is ready. Like the other replies mention, we try to measure how long food prep takes for a restaurant and we may err on the side of caution with restaurants with high variance by sending them to a restaurant assuming the shortest estimated food prep time. * Restaurants don't always provide accurate estimates of how long the food is going to take to make. * In your specific anecdote, we likely predicted the food to be ready much earlier (atleast earlier enough that you wouldn't have you felt they could have done another delivery) but some things may have changed from the original delivery. Eg: The kitchen was especially backed up during this time and we didn't have examples of this in our dataset. Or perhaps we did have examples of this and it didn't happen enough to increase its weight in the food prep calculation. Customer may have made an adjustment to the order, which caused additional wait time after we made the original dasher/delivery assignment. Kitchen may have made a mistake with the order and had to remake it. Our estimate for drive + parking time may have been too high. In other words, some of the time that you saw the dasher in the restaurant could have been time that we estimated for driving/parking instead. There are a ton of other factors at play, but I hope this gives you some sense of complexity of the problem :)

danieltillett 11y ago

This is the exact thinking that got all these companies and investors into this niche. Unfortunately, the restaurant business is too messy to automate (at least right now) and too prone to falling into profitless competition. There is a great business in this area waiting to be unleashed, but I have no idea how anyone can make a profit.

jimkri 11y ago

The "dashers" could also keep track of the wait times for each restaurant, and then work accordingly to the wait times. They could boost their delivery's, but it definitely would take several weeks to get the right amount of knowledge about each restaurant.

mathattack 11y ago

The listed wait times on the App are 60-70 minutes. As a former New Yorker (used to bike delivery of sub-30 minutes) it seems crazy that they could have a viable business. It's even more surprising given that they don't have food warmers, so the food arrives cold. This tells me that they're not really in the Food Business - they're in the delivery business, and learning how to do it with food.

potatolicious 11y ago

The trick with DoorDash is that many (most?) of the restaurants in the system aren't voluntarily participating with DoorDash. A smart system that summoned delivery people when the food was ready/close to ready would require integration on the restaurant's part (see: Seamless/GrubHub/Caviar), but DoorDash seems to be going in a different direction. This has been a point of controversy - DoorDash offers delivery for restaurants that do not participate in delivery by simply sending someone there, ordering, and bringing you the food. This is terrible for the delivery people obviously, and DoorDash has also been caught marking up the prices on these menu items without informing the customer. Overall, highly doubtful about the sustainability of this business model. Or the ethics. Side note: Postmates also does similar - some of the restaurants in their system are not official partners, and so the delivery person gets to make the order and wait around. This is a large part of why I stopped using them - they're getting paid peanuts as it is and now they're sitting on their ass for 20+ minutes to bring someone $20 of food and foregoing much needed tips.

bhahn 11y ago

(disclaimer: I work for DoorDash but have minimal exposure to our scheduling platform) > it would have been more efficient to schedule other deliveries in the mean time and just come pick up the food when it was ready? There are a lot of factors at play so it's hard to say what exactly happened in this instance, but some considerations: * The wait time at a restaurant is probably not enough time to fit comfortably another delivery. Deliveries take a bit of time. I don't have the exact #'s, but you can imagine 5-20 minutes driving to restaurant (+ parking), 0-5 minutes waiting at a restaurant, 5-20 minutes driving from restaurant to customer, which ends up being around 10-45 minutes with high variance. * We want food to be as fresh as possible for the customer, so on average we may prefer a dasher to wait a little rather than arrive the food is ready. Like the other replies mention, we try to measure how long food prep takes for a restaurant and we may err on the side of caution with restaurants with high variance by sending them to a restaurant assuming the shortest estimated food prep time. * Restaurants don't always provide accurate estimates of how long the food is going to take to make. * In your specific anecdote, we likely predicted the food to be ready much earlier (atleast earlier enough that you wouldn't have you felt they could have done another delivery) but some things may have changed from the original delivery. Eg: The kitchen was especially backed up during this time and we didn't have examples of this in our dataset. Or perhaps we did have examples of this and it didn't happen enough to increase its weight in the food prep calculation. Customer may have made an adjustment to the order, which caused additional wait time after we made the original dasher/delivery assignment. Kitchen may have made a mistake with the order and had to remake it. Our estimate for drive + parking time may have been too high. In other words, some of the time that you saw the dasher in the restaurant could have been time that we estimated for driving/parking instead. There are a ton of other factors at play, but I hope this gives you some sense of complexity of the problem :)