Dumpling Express is one of four businesses that uses the TooLow app to offer deals to its customers. Photo: Sarah Han

Enough food delivery apps have come out of Berkeley to fill up a smartphone screen. But a new local food app is delivering savings on food rather than delivering actual meals to its users.

TooLow is a free mobile app that allows users to purchase food from local restaurants at a discounted price. TooLow does not deliver meals; customers still have to pick up their orders in person, but by using the app, they can get the menu items at participating restaurants for less than the posted menu price.

A banner at the top of the app’s homescreen tidily explains why a handful of Berkeley restaurants have decided to partner with TooLow: “Food prices drop so restaurants can sell their inventory faster.” At the time of publication, those restaurants are Nefeli Caffe, Dumpling Express and Abe’s Pizza, with occasional “guest appearance” style bargains from other stores such as the Chipotle on Telegraph Avenue, Sharetea on Bancroft Way and Gypsy’s Trattoria Italiana on Durant Avenue.

TooLow was not originally conceived with food in mind. The company’s CEO, 21-year-old entrepreneur David Akerman, transferred to Cal from the London School of Economics in January 2015. He was only enrolled for one semester at UC Berkeley before leaving to work on TooLow full-time. He launched the app in San Francisco as a way for consumers to find deals on haircuts, but he quickly changed both its focus and location when he realized he wasn’t going to get enough repeat users at the frequency needed to keep the business going. That’s when TooLow switched to food and he brought the service  to Berkeley.

It is a truth universally acknowledged that college students order takeout. And with Berkeley’s history of support for nascent mobile food apps — Caviar, Doordash, Sprig, Copia, Spoonrocket, Munchery, Hopsy and Kiwi have all come out of Berkeley — the city and the campus made for an ideal testing ground.

“Students seem to like the fact that it’s cheaper and easy to use,” said Akerman, though he has been seeing some customers who live in Oakland and even San Francisco, but who might be passing through Berkeley and using the app to get a deal on lunch.

But since launching in April of this year, TooLow’s usership has grown from an initial pool of around 25 to more than 2,000, 70% of which are students and, unsurprisingly, mainly centered around campus.

The basic idea of TooLow is a “reverse auction.” In a standard, or forward auction, prices steadily rise as individuals outbid each other to purchase goods or services. In a reverse auction, the roles of buyer and seller are flipped, and pricing steadily lowers.

In a standard reverse auction, businesses with a extra commodities compete with each other, trying to undercut competitors by lowering prices until a buyer says “Yes.” Essentially it’s a race to the bottom but wherein both parties benefit.

Where TooLow innovates is by shifting the concept from the old market — large quantities, traded business to business — to the new one — small quantities, traded business to consumer.

“This concept was never applied to food,” said Akerman. “This app is the first time ever.”

On TooLow, partnering businesses set floor and ceiling prices. Then an algorithm determines a figure within those parameters to offer the optimal price for the current market. It works similarly to rideshare surge pricing, but lowering, rather than raising prices. TooLow takes a 15% cut on every purchase made.

It might seem counterintuitive that a restaurant would drop prices to make money, but there’s a simple reason why. “Physical is slow,” said Akerman. “Long lines at prime time can lead to less sales.”

As a line gets longer, fewer customers are inclined to jump on. Which means that a restaurant can lose potential customers both in spite and because of its popularity. Offering deals via TooLow means being able to serve more customers, faster and more efficiently. In the world as presented by TooLow, a restaurant could make even more money at the lunch time rush, even if they’re selling salads for 44 cents below the menu price.

The majority of savings are in the range of twenty to fifty cents. Though there are often deals before or after the lunch rush where a user can save much more.

Assistant manager Peter Chan of Dumpling Express acknowledges that TooLow has brought his restaurant an increase in sales, though only perhaps an additional eight to ten orders a day, depending on that day’s menu. “Sometimes we have specials on particular items so we sell more of those [through the app].”

As a whole though, Chan is pretty measured in his own enthusiasm. “They’re helping us a little bit,” he said.

Not all restaurants partnering with TooLow post all of their items, but they do tend to post the most popular. Which is also why users have to jump on deals while they last, or else risk seeing an item followed by two words in all caps: ALL SOLD.

A screenshot from recent discounts on the TooLow app

With TooLow, both the food items and the users are evaluated by a price. Items are assessed in dollars, while users are measured by a unique 12-digit number. Or as it’s called on TooLow, the “Human ID.”

The idea behind the Human ID is to be a shorthand for an individual’s buying preferences and spending power. Currently, TooLow uses it as a way to track when users are looking at the app, when they buy and at what prices. The app can then alert them based on their preferences. Perhaps a particular user won’t click “buy” for a 15-cent discount, but will for a 30-cent discount. In which case TooLow would record that preference and not send an alert until the preferred item goes down in price by at least 30-cents.

But Akerman doesn’t just want to stop at alerts. He also wants to use what economists call “fuzzy pricing,” which means creating different prices for different customers based off their preferences and ability to pay. Think of it like automated haggling, but more rational, not just accounting for a user’s emotions or desires, but — eventually — also for their current financial situation.

“We want to solve inefficiencies in the economy,” said Akerman. “One of the biggest inefficiencies is pricing. People don’t know how to do it. But if we can price everyone correctly,” Akerman continued, “we can close the gap of wealth inequality.”

It was at this point that Akerman pulled out another economics term: deadweight.

In economics, deadweight loss happens when a good or service is incorrectly priced for the customers it’s trying to serve. Maybe Tony’s Panini Palace can count on Daddy Warbucks to buy a $10 panini every day. But if Daddy Warbucks is the only customer able to afford a $10 panini, then all the unsold paninis leftover at the end of the day are wasted. That’s economic deadweight.

Tony’s could lower the menu price to $9.70 so Little Orphan Annie and all her friends can now buy paninis, but now the restaurant is losing 30-cents on the panini for every customer like Daddy Warbucks who can afford to pay the full $10 price. That’s also deadweight.

Which is where an app like TooLow has the advantage. Tony’s can’t change the prices on a standard one-size-fits-all menu posted above the cash register. But it can change the prices sent as individual alerts to people’s phones, and with all the privacy of a text message. Which means neither Daddy Warbucks nor Little Orphan Annie has to know what anyone else is being charged. They just have to know their own price.

At some point, Akerman hopes all people will get some form of Human ID. And that it can serve as a kind of shorthand identification in the way that social security and driver’s license numbers currently serve.

“We don’t just want it to be as important but as ubiquitous,” said Akerman. “We want it to exist whether TooLow is there or not. We want it to be something all seven billion humans are going to have.”

Akerman imagines a vast and virtual auction space where algorithms match users with prices, and where products find their buyers. Where reverse auctions and fuzzy pricing can eliminate deadweight, improve the global economy and reduce inequality.

Eventually, Akerman wants the technology behind the app to put an end to global poverty. But for now, at least, he’ll settle for saving college kids a couple bucks on burritos. Akerman has plans to expand TooLow to 30 different college campuses by the end of 2018. Next on the list are UC Davis, the University of Tennessee Knoxville and Harvard. Like so many other ideas, today it’s Berkeley, and tomorrow, the world.

Nosh Weekly is a free weekly email that will keep you bang up-to-date on all the delicious food, drink and restaurant news in the East Bay. Sign up here.

Cirrus Wood is a freelance writer and photographer living in downtown Berkeley. There are few things he enjoys as much as playing around with the alphabet.