Hacker News Reader

Comment by sharkmerry | Hacker News Reader

> 2) Gross Margins (GM). As we grew, our burn rate grew too. We were losing money on every meal. If only we could get to critical mass.

> We had epic revenue growth w/ burn rate growth. Soon we were burning $1.5-2M/mo!

> We were always “1-2 months away” from managing the burn.

I dont understand, how in good faith, he can bring up SF-regulations as issue #1. When issue #2 is the crux of all their issues. They lost money on every sale! Food industry is notorious for thin margins, and then had negative margins and growth. SF regulations dont matter compared to that.

View on HN

People who are not in the actual retail restaurant industry do not understand that there's no way to make money with ghost kitchens. Theoretically, to make money with a ghost kitchen one needs it to be churning out a very large number of standard meals at a price that customers want to pay to achieve the economy of scale which at the same time requires an economy of scale in logistics, which has its own overhead. By the time you figured out how to make eick out money there, customer's preference changes and you need to start from scratch or a competitor shows up and cuts your margins.

> a very large number of standard meals

Or a number of meals all made of the same 5-to-7 ingredients arranged differently, i.e. the McDonalds / Taco Bell model.

McDonalds/Taco Bell do not have ghost kitchens. If all of their meals were completely and totally made off site and were delivered to the customer to order from an off-site then those would be examples of ghost kitchens.

I didn’t mean to imply that they did; just that their kind of menu-planning approach—of having many different (fast) final preparations that all take as input the same small set of pre-cooked ingredients—would also be a workable model to copy, to achieve the economies of scale that ghost kitchens require.

ghost kitchens can absolutely make money. Them being done at scale with VC money, most likely not.

But as the below commenter said, if you can use same ingredients across multiple dishes, it reduces waste and can reduce COGS. A small restaurant can do it, the issue there is establishing a brand, but if all your orders are coming via 3rd party apps. not sure how much brand awareness you actually need

> ghost kitchens can absolutely make money.

That is as assumption that has been made time and time again and it always failed. The most important part of a successful kitchen is a near zero dwell time. The smallest dwell time is achieved by having a kitchen be colocated with the customer.

As someone who has never worked in the restaurant industry, I don't understand, at all, what you mean by this. Could you elaborate?

Not them, but I imagine that it's because your fixed costs never stop — you can't just tell your staff to go home and not get paid until the next customer gets in.

Every hour you're open, money goes out. You need to have continuous sales to even have the opportunity to make ends meet. Even if you're losing money each hour but still making a few sales, you have some flexibility with tweaking your formula to reduce the losses or turn it into a net even. If you're just sitting idly, you're just burning money without anything to do about it.

Dwell time is the time between the order is finished and it reaching the customer who gives thumbs up or thumbs down ( order is wrong ).

It seems from the initial glance that dwell time is irrelevant because a kitchen can just go to the next order but that's not actually the case because without yes/no from a customer the kitchen can continue to make the same errors without adjusting. In the world of ghost kitchens dwell time is terrible because in reality the only feedback the kitchen gets is an asynchronous negative review that will be received a day or two later at best. That's the reason why pretty much all ghost kitchen based restaurants rapidly flame out.

"ghost kitchens" that make money are called factories. Factories where they make frozen pizza and lean cuisines.

Reply on news.ycombinator.com