[Mitchhhs]

"Xu said the company became “contribution margin positive” in the last year, which means that it’s profitable on a per-order basis. In fact, DoorDash has become profitable in its earliest markets."

Its funny this is a milestone for a company at this stage. Congrats you aren't losing money on every order!

[WisNorCan]

> Congrats you aren't losing money on every order!

I think this misunderstands how marketplaces are built. Traditional economics would agree with you and say that selling $1 at $0.90 is insanity. Traditional economics have been slow to catch up with marketplace economics although there have been some recent papers that start explaining a better framework.

If you know that a marketplace has strong network efforts and improved performance at scale. The right strategy is to understand the trade-off between time and $ required to get to scale in your marketplace.

The optimal strategy often means subsidizing a market at a price below your actual cost for early markets. The fact that Door Dash talks about their business in terms of early vs. late markets with different economics means that they are using this playbook.

There are of course many companies that think they are building a marketplace with a network effect when they are actually just losing money. I don't know DoorDash specifically.

[ghaff]

I'm not sure I even understand what the network effects are here. I guess there are some potential efficiencies with having a single delivery company that handles deliveries rather than individual restaurants but it seems like a stretch. People I know who live in cities tend to just have a stack of take-out pizza/Chinese food/etc. menus for places that deliver (or not). Maybe I'm missing something but it doesn't really seem like an area that's crying out for "disruption."

[fnovd]

Using DoorDash or UberEats as your delivery service instead of building your delivery staff in-house is like using AWS or DO instead of building your own in-house ops team. Since most delivery workflows are pretty much identical (go to restaurant, pick up food, deliver food), it seems like any restaurant other than a "typical delivery" place would benefit from using someone else's infrastructure.

[ghaff]

Except that "building your delivery staff in-house" more or less consists of hiring some teenagers to work for tips. It's not exactly architecting a datacenter. Furthermore, delivery is going to inherently be a local business even if you're a nationwide company.

I actually do think that there's something to be said for having a standardized service that a restaurant already offering take out can just sign up for. Maybe delivery is something they just never got around to offering. It just seems like a service that's hard to do well at a price people are willing to pay.

[capsulecorp]

Maybe its where you live? I live in Las Vegas, and with the 24 lifestyle that often includes drinking, these services have absolutely blown up out here. Granted its a pretty special case in terms of city density coupled with a larger than normal group of restaurants that are open late or 24 hours. That and the 215 beltway makes it so even if someone wants something from across town its not much of a hassle (lest its during rush hours obviously). I always wondered if it was the same in other cities, I just don't see it as sustainable without the 24 hour life style and the city density.

[ghaff]

It's fair that there's been an increased expectation of near-immediate gratification. As a result, although a fair number of restaurants in cities have long offered delivery, it's not clear that ad hoc delivery on a restaurant-by-restaurant basis necessarily scales to this new world.

It will probably vary by location though. I live about 40 miles out of Boston and adjacent to a couple smaller cities. I have basically no food delivery options.

[seizethecheese]

Restaurant delivery people will be utilized 50% (trip there and back). If the network effect can do better than that, then there's a business.

[wuliwong]

I'll see your anecdote and call with mine: I live in Atlanta and I can't think of a single person that doesn't use some app to order delivery anymore. Do you live in a city or just know people who live in cities.

[Mitchhhs]

Yeah that's the problem here its really hard to tell the difference between the two. Sometimes it feels like a "new economy" excuse.

Question, what is a good example of a market place that is now highly profitable where the unit economics/contribution margin were negative for a very long time.

Amazon is not an example, I guarantee they were making contribution margin on every book sold fairly early on.

[not_that_noob]

Investors found that particular religion with Google. Losing money hand over fist till they put in adwords. They had to fight investors to put in one of the most profitable models on the planet. Ever since then, the game has changed. Investors now believe fervently, and founders realized investors may not know everything. Everyone else who came after drafted in Google's wake (I'm looking at you Zuck)

[Mitchhhs]

Yeah but thats different. Theres a difference between charging zero to get a network and losing money as a business vs selling a product and losing money on every single unit of that product being sold.

[not_that_noob]

Basically, it's like Indy Jones with the leap of faith - there's a an inordinate amount of hope in the future.

https://thumbs.gfycat.com/PoorImperfectAdouri-size_restricte...

Thankfully, the world is afloat with cash looking for a home, and so we can continue with this game for a while. At least, I think the portal will stay open till the round closes. Which is enough mostly.

BTW, traditional economics can well model this using advanced options equations, but the answers won't come close to justifying the valuation they got.

[teambljck]

Possible to share link to said papers? I'd love to get up to speed on the first principles behind the frameworks

[notyourday]

Creative accounting at its finest.

Company is either GAAP profitable, pro-forma profitable, EBITDA profitable, cashflow profitable, cashflow neutral or money losing. All other metrics are creative accounting that put lipstick on a pig masking money losing status.

DoorDash is money losing.

[bfrog]

Which is amusing given they don't have to pay the actual labor to deliver goods or make food. Are they set up in the valley paying mega wages perhaps?

[ac29]

Not sure why you think they don't pay their drivers. The exact amount is unclear, but according to this FAQ [0], it is "at least $1" per order, and at least a variable "guaranteed amount" that door dash will pay up to, regardless of customer tips.

[0] https://dasherhelp.doordash.com/new-dasher-pay-model-faq

[thesimon]

>> became “contribution margin positive” in the last year, which means that it’s profitable on a per-order basis.

>Congrats you aren't losing money on every order!

Seems like an overstatement as fixed costs are still not covered when they are "contribution marigin positve".

[apendleton]

Definitely. At least, though, it means there's a path to profitability just by increasing volume, right? Like, if every new order brings it more than it marginally costs, with enough eventually you'd cover your fixed costs, so it seems like at least a rough validation of the business model (assuming their total addressable market is as big as they think it is). Vs, say, Uber, which is as far as I know still losing money on every ride and has no path out aside from not at all certain, majorly disruptive technical innovation.