To our faithful customers and partners,

You’ve probably heard of Radish through a friend, your extended network, or from one of our restaurant partners. You may be vaguely aware of the fact that we are a restaurant industry cooperative, or that we seek to increase revenues and lower costs for our restaurant, delivery and consumer members. You might even have signed up for our online delivery platform and enjoy our service. The site works, though you’ve probably found that it’s missing a couple of nuts and bolts and what would seem like some pretty basic features, like a homepage.

The truth is, we don’t have a fully developed homepage to show you. Just like we don’t have an app. And although we’re a bunch of software engineers and data scientists, our back office is actually comprised of us greasing phone calls to delivery drivers and ad-hoc accounting and inventory on shared Excel docs. Actually, we’re usually the delivery drivers too.

I know the typical narrative is that a group of young upstart entrepreneurs hack together a killer app in their garage that unseats the market incumbent in a stunning upset. Startups are supposed to have a superior product and technology that they can iterate on faster than their comparatively sluggish competitors, and hopefully this is something we will have in due time. As the situation stands, however, we simply don’t have the resources to cross the ever-widening technological moat being dug by Big Tech. Virtually no angel or institutional investor in Quebec would be able to compete in terms of the dry powder that the mega-funds and sovereign wealth funds have supplied to prop them up — at least not on nominal terms. The thing is, even if we could get our hands on such a war chest, we’re not exactly convinced that it would be worth it.

So then, how do you outrun a unicorn? One with infinite stamina that allows it to sprint at the speed of light and mythical powers that allow it to distort realityas it sees fit?

How about we first ask ourselves: why outrun a unicorn?

Have you ever considered the breakdown of what goes into a restaurant meal? You know, in the Financial Times, cup of coffee with 1 cent going to Colombian growers' kind of way?

It's really difficult to obtain reliable statistics on this[9] thus I'll try to illustrate it with napkin figures from my prior restaurant experience.

The first 27-35% goes to the food costs. This varies by restaurant concept and by food type. This is everything from your purchase of brioche buns and beef patties to garlic, oregano, mustard, baker’s yeast, etc., but also your spoilage, theft, improper inventory accounting and the like. Approximately 30% of the costs are labor. Again, this varies a lot by concept. For some fast food restaurants, this can be several points lower and for many higher end establishments, this can hover above 40%. Stellar benefits are not the norm in the industry and for many employees, they may consist more of unconventional favours, such as the customer’s mistaken margherita, cash advances on the week’s pay, or the owner playing a helping hand in immigration paperwork.

Combined, these percentages represent the brunt of the prime cost. This is basically the production costs of providing a meal to a client. This is arguably the most critical cost for a restauranteur and essentially sets the menu price of an item.

The next largest chunk is your fixed recurring expenses such as rent, utilities and insurance, which can easily take up 12 to 15%. This sets the lower bound for the revenue necessary to stay afloat. You then have variable operational costs such as supplies (cleaning products, custom packaging, printer paper, aprons, delivery bags, coffee stirrers, etc.), marketing (paper and digital), services (commercial music licenses, uniform and tablecloth laundering, knife sharpening, floor waxing, bookkeeping, etc.), equipment maintenance and miscellaneous, which make up 10 – 16%. Being variable, this can deviate significantly between restaurants.

Payment processing takes approximately 2.5 – 3.5% for every credit card transaction. Fees are on the higher end for online channels due to the higher fraud risks for payment processors. If the restaurant is a franchise, they may have to pay a royalty of 3 – 9% on gross sales, though this, in theory, is supposed to defray some of the other costs such as marketing.

If you've been doing the math, you should have come up with something above 90% as an average.

We’re not quite done yet. We have to account for the amortization and depreciation of all the equipment, namely ovens, range hoods, walk-in fridges, mixers, coffee machines, plates, cutlery, printers, computers (with POS software), tables and chairs. It’s hard to put this as a percentage, though the dollar amount for this investment is normally in the six figures (the tail ends being 5 and 7 figures).

Virtually no one can pay for all of this out of pocket, so some form of financing is a given. Large equipment and real estate can be collateralized for a more favourable interest rate, but cash flow for operational expenses is generally managed through credit cards, lines of credit, private lending and IOUs to friends and family . We thus figure interest payments as a line item.

All things considered, we’re left with 3-5% as a pre-tax profit for the typical restaurant. Along with grocery stores, this is amongst the thinnest in any industry.

And if all goes well and you’re able to move it up to 8% or 10% or even more and you can manage to do 7 figures in sales… you’ll be rich.

You'll be rich. Still in debt of course, but rich. (And then you'll promptly move into the franchising business to longer have to deal with these economics).

Rich.

But one month, your lease comes to an end and your landlord, seeing the steady flow of clients or perhaps faced with their own city tax increases, hikes your rent. Or the city has to do aqueduct work which blocks your storefront for the summer months. Then a motorcyclist gets hit there in the fall. Or there are trade tariffs imposed on tomato imports. Or the chef who does your weekend shift gets deported. Or your restaurant moves to the 2^nd page of results on TripAdvisor. Your profit margins fall to 1%. Or -1%, or less…

Mix and match for maximum magnification. If you’re fortunate and disciplined, you have the cash reserves or credit to last a few down months while your business builds back up to profitability. The more likely scenario though, is that you start making cuts on various aspects of your business in a desperate bid to make ends meet. You hire a more affordable, but less experienced chef, use lesser quality ingredients, and charge more. Clients are now having to pay more for less. Business declines even further. At this point, your restaurant starts a vicious death spiral and it becomes very hard to salvage the situation without external sponsors.

It should come as no surprise that something like 60% of restaurants don't last 3 years and 80% shut their doors by year 5. No bank can stay solvent playing those odds which is why traditional loans for restaurants are uncommon and costly.

Now let me ask you, when you order a meal through a platform like UberEATS, DoorDash or Deliveroo, what do you think the restaurant pays? 10%? 15%? 20%?

Well, on average, the cut will be 25% to 35% (though this has gone up to 40% in certain regions).

This is not as terrible it sounds. In theory, you’re not supposed to be factoring your dining room costs and to a certain extent, marketing, and since you’re relying on the provider, most of the logistics of delivery are externalized as well.

As discussed earlier, however, with its low margins, the restaurant business is hyper-sensitive to costs. If you’re a restaurant that is doing 5000$ – 10000$ of sales on a delivery platform a week, the difference between 25% and 35% is basically the wage of one or two staff members.

Despite this, if this was the entirety of the calculation, the economics would still be tolerable. What remains to be factored in are the ongoing costs associated with the right to sell on the platform. And no, this isn’t some fixed fee that you have to pay, that would actually be preferable (though of course, you may still have to pay this through the tablets and installation fees).

See, the platforms have gradually captured more and more of the mindshare of consumers, especially millennials, becoming the sole way many people will even order food. As such, the platform can decide which restaurants gets showed to the consumer and which… get showed less. Fortunately, the platforms are kind enough to inform you when you’re not getting a lot of visitors. They’ll tell you that investing in marketing will help you drive more people to order from you. They’ll have some helpful suggestions, like offering 50% off on popular products or covering the delivery fee for the customer. They’ll pitch in, by promoting you on the front page of their app. The next day, your restaurant is discovered and your orders quadruple. Lo and behold, you’re the most popping restaurant on the street! I mean, at least until everyone else invests in marketing, right?

The day Foodora mandated marketing for the local industry as a whole

You’re not obliged to continue marketing forever no doubt. Eventually you stop and your newly loyal customers keep coming back to your restaurant looking to pay 100% more than what they what used to pay. Just kidding. Your restaurant goes back to where it belongs, which is page 5 of Caribbean, Greek, Pizza or whatever category you’re in.

Why not just continue? Perhaps not with the 50% off, which is admittedly, a bit of an extreme example, but with just footing the delivery fee? If DoorDash is taking 35% of your 20$ order of green chicken curry and papaya salad and then you pay another 3.99, all of a sudden, you’re left with less than half of the order. A harder pill to swallow.

I could stop here, but it’s really not the most egregious behaviour of the online delivery platforms. The aforementioned issues are still local and human in nature. Meaning the savvy restauranteur can understand them and try to negotiate or adapt on an individual level. The larger issues are systemic and harder to perceive unless you’re actively looking at the bigger picture.

I think it’s safe to say that all the online ordering platforms have as a mission to provide the best marketplace for their users. Now, what would that look like? What if they could create the perfect platform where the best restaurants would be serviced by the best delivery drivers and cater to the best customer? So, like, if you were a customer, you would be showed the highest rated restaurants in your area and once you ordered the nearest driver would quickly bring it to you. Inversely, as a restaurant, you would get larger orders and those orders served promptly and as a driver, the nearest routes and customers that tipped more favourably.

As you can imagine, this is a bit unwieldy for humans to manage on their own. In 2017, UberEats claimed it had over 46000 restaurants and in 2018, they announced that they were in 200 cities and were opening in 100 more. You thus have algorithms that take in numerous factors such as restaurant ratings, driver location, customer order history and optimize everyone's interactions on the platform to create the best-case scenario for all the parties.

We keep using the term best. But what is the definition of best? Not to us, but the platform provider? ? It would be ideal if there was a way of assuring this meant restaurants that had amazing food, superior working conditions and put a lot of passion into their work or delivery drivers who were fast and offered great service and customers who tipped generously and thoroughly enjoyed their experience.

Unfortunately, algorithms don't innately take cultural, societal or ecosystem concerns into play. They take whatever biases their owners impart, knowingly or unknowingly. And in the case of the large platform providers, the initial bias is optimize for profitability.

But Uber processing a higher volume of orders means more revenue for merchants, right? Not quite. Profitability for Uber is not necessarily profitability for restaurants, or for couriers, or cost savings for customers.

To demonstrate, imagine a hypothetical online delivery platform with three fried-chicken joints serving a geographical area: Restaurant A – an established family run business, Restaurant B – a newer establishment by an up and coming chef and Restaurant C, a greasier chain restaurant. In an ideal scenario, with all else being equal, the three restaurants might receive roughly the same promotional coverage and similar prioritization by drivers, or depending on your platform’s policies, maybe a bit less for the shabbier Restaurant C and a bit more to help Restaurant B build reputation on the platform. Drivers would be distributed based on the demand from customers for the various restaurants.

This would be one of the more equitable ways to operate, but this would not necessarily be the most profitable outcome for the platform. Let’s say that because of the rate it negotiated with restaurant C and the size of its orders, the platform makes 2$ for every delivery it makes for them vs. 1$ for the others. The platform has several levers to generate more business for C. It could constantly promote it on its front page, proffering its proprietary rating as being a signal of recognition to consumers on the platform (ever wondered why Skip has a Skip score?). It could keep more drivers around restaurant C, incentivizing consumers to order from them, since the delivery times are considerably faster (ever wondered why Uber’s free delivery offers always cycle through the same restaurants?) Restaurants A and B start getting less orders, until of course, they’re approached for marketing spend that happens to conveniently equalize or surpass the returns with restaurant C for the delivery platforms.

Does this sound remotely problematic at all? The platforms can essentially decide the winners and losers. I mean, they might as well start their own restaurants using all the data they’re collecting and effectuate their soft power to influence consumer purchasing trends towards their own first-party channels. Oh wait…

https://www.restaurantdive.com/news/uber-eats-serving-up-virtual-kitchens-in-paris-bloomberg-reports/550398/

I don’t want to paint this as some sinister conspiracy. There are a lot of good engineers and researchers who are trying to build more ethically robust systems. It’s just that this is very hard to do, both from a technical standpoint and also because shareholder interests are the ultimate starting bias to any algorithm. How can we really verify that these algorithms try to keep anything fair at all, when there is little transparency and the companies have consistently shown themselves to want to do anything but?

We haven't even broached the topics of the platforms charging 20% - 35% to delivery drivers who risk making less than minimum wage and then charging 3-7$ to the client for delivery fees. I could keep going on and on.

Where did we go wrong? In virtually every industry, we are told that technology is supposed to be a great enabler and force multiplier— Lower prices and higher profits due to increased productivity. Computers made it cheaper than ever to consume and produce entertainment, communicate with a large audience, optimize and manage supply-chains, perform drug discovery, make it easier to start a business and millions of other things.

But why then is it that in the restaurant industry, it seems like technology has squeezed us?

And what happened to age-old industry paradigms such as 30 minutes or its free? Pizza and Chinese restaurants were doing free delivery for decades before online delivery platforms came into the playing field.

No one is winning, profitability is far from being a given for any of the larger players. What blood is there to extract from a rock by taking 40% from an industry whose profit margin is so small that almost half of those who work in it live in destitution?

To add insult to injury, this is to support the salaries of thousands of engineers who earn six figure salaries and live and work out of the most expensive ZIP/postal codes in the world. These are the business models we want financed by our pension and tax money?

The reason we’re not rushing to cross the technological moat because is because it’s dishonest to pretend that there is an app or AI solution that will make the lives of people in the restaurant industry in better. Tech is not a panacea. It is not an end, but a means to other forms of societal innovation. Whether that be Facebook, Tinder, AirBnB, Uber, Amazon or Microsoft.

And in the case of the restaurant industry, the real innovation of all these platforms is the shifting of the economics and social contracts between the actors in the restaurant ecosystem, namely those between restaurants, their staff and their customers. These contracts that used to be between communities of restaurants, staff and customers have been eroded and supplanted by ones with a few large technology players that end up dictating the terms of the market. It’s not all bad— change is important. There are, however, some very important concerns here.

Not too long ago, I saw an article on the onset of the coronavirus epidemic in Philadelphia. It said something like 500 000 people in the hospitality industry were at the risk of losing their jobs. I wonder how many were able to hold on.

In Quebec, that figure is something like 270 000 people. As of over a week before the publication of this post, over 175 000 of them have been laid off.

It breaks my heart. For the longest time, it was the industry I called home.

My parents owned at different times, several franchise and independent restaurants. Until a few years ago, I had spent more summers than not, cooped up in an unventilated storeroom closet that we used as an office. I’d sit there, waiting to see mom after her marathon 12-hour shifts waiting tables. Lucky occurrences were bedtime stories, when per chance my dad would be home before I fell asleep.

By the age of 9, I joined them. At first, it was mainly bussing tables and getting in the waitresses’ way. Soon after, I was waiting tables and giving a go at everything else. The waitresses weren’t very pleased with me taking their tables, but they still loved me. Even though they worked paycheque to paycheque, they still always got me presents for Christmas.

At 15, I was running my own restaurant, Côte-Des-Neiges Hot Dog, that my dad helped set up in an unleasable commercial space. I had two employees, a dishwasher who was drafted as a cook and girl who wanted a first job so badly, she literally sat outside all day, waiting for the manager to come back. By 16, my landlord (aka Dad) kicked me out because he found a tenant who could pay more. He was generous enough to hire me back however.

Humble Beginnings

Some days, I’d get locked in the walk-in fridge playing hide-and-seek by the older brothers who worked in the delivery staff. Some days I’d sneak into the walk-in fridge and eat all the brownies.

I have so many fond memories with the chefs. They served so many of my meals and spent so many hours with me that they pretty much qualify as my parents. Their own children are now doctors and engineers. Their retirement plans consisted of opening their own pizza places banking on the creditworthiness of their educated offspring. It sounds a bit off until you think of what their investments really were.

The restaurant industry is diverse, one with low levels of education and astonishing poverty, one with incredibly passionate and creative individuals from all walks of life. Some are supremely educated, some are famous, some are secretly rich, some are addicted to substances and everyone is convinced their business partner is out to get them. Everyone loves the dishwasher. Waitstaff hate their customers, but they’re their only family. Managers and their staff hate each other and are also their only family. Family succession feuds are responsible for more M&A activity than any private equity investor. The pay is trash, but what other job lets you walk off a Friday shift with 500$ in cash?TGIM.

In all this contradiction and uncertainty, there is one thing that is constant, which is that the restaurant industry is brutal. People love what they do, but everyone who goes through it comes out a bit broken. Sometimes very broken.

I spent the last several years working in technology and in finance. I’ve never made so much money in my life (except maybe working the Friday shift alone). And I love what I do now. However, as far as I go, I find it hard to stay away. There is a thrill and a kinship that I haven’t found doing anything else. More importantly, a lot of people I know are still working there and trying to make ends meet.

This is why my friends and I started Radish. We’re building an online delivery cooperative, which is an organization where the core constituents, restaurants, staff and consumers are all part owners— literally. We believe it’s the only way that such a critical piece of infrastructure can be managed fairly and while respecting everyone’s place in the ecosystem.

We’re not sure if this is going to work out, but if it does, it’ll return ownership of the technological means, the algorithms, the data, and intellectual property back to the people and businesses who rely on it for their livelihoods and their passions. This is not a charity; we are a for profit venture, but the profits are to be redistributed to the industry where they rightfully belongs.

So, to answer my question from earlier, how do you outrun a unicorn? One that can run as fast lightning and soar through the clouds? One that can use its magical powers to remake the race as it sees fit?

You stop believing it exists.