Hacker News new | past | comments | ask | show | jobs | submit login

An accurate estimate isn't really possible with the food delivery business model. Couriers will have 2-3 orders, so if you are order 1 or two, there is a new variable that enters the equation after you purchase the food. They obviously add other items on a similar route, but wait times and such are fairly random). Of course they could do 1 order to 1 driver but that would double the labor costs (sequential orders would be 2-3 deliveries per hour max per courier, vs 4ish when optimized).





That doesn't explain why the delivery consistently jumps 30% to 50% as soon as you hit the confirm button and they process your payment. If they were giving an honest estimate, you'd expect a gradual increase in the estimate as the delivery person showed up 2 minutes late at point A, then 5 minutes late at point B, then 10 minutes late at point C, etc.

The Uber Eats delivery time estimates are a lie, plain and simple. Once they have your money in hand, they'll shamelessly admit to the lie, which is why the estimate jumps instantaneously.


also, you would expect to at times get over-estimated wait times. but my experience is always underestimated by about 20 minutes. so consistently, in fact, that I suspect they have a very good idea of the true time and then subtract 20 min from it.

Then they shouldn't charge for faster delivery. It just doesn't make any sense. Like i'm 95% sure in most cases you don't actually get a car faster and they just take your money and provide no marginal value (granted, I only use the apps about every other month). How else could they consistently offer multiple price tiers regardless of availability? I'm not claiming fraud necessarily, but I also can't imagine how you could offer it without committing fraud. At some point you're just lying ("estimation with negative effort" as a friend calls it) about car etas to drive the perception of value.

The estimate is not really that accurate even when you pay the $4-5 more to be the first delivery.

Uber could provide a range or a point in the middle, but it seems like they just estimate the best case scenario and then it's usually not the best case.



Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: