Most mobile order-ahead platforms have one simple goal: Get meals to consumers’ doors as fast as humanly possible. Sometimes inhumanly fast is okay, too.
But DoorDash believes there’s a better science to perfecting the speed of food delivery. The San Francisco, California-based company is using machine learning to make sure meals stay hot, whether they come from a restaurant down the street or across the city.
“We’re a last-mile logistics company that is focused on food delivery,” DoorDash product manager Abhay Sukumaran said in a recent PYMNTS interview. “What we’ve wanted to do has always been [to] figure out the best way to help local businesses reach more customers through delivery … and one of the things DoorDash has always focused on is how do we bring the best selection of restaurants in your city to you [the consumer].”
The company connects three different parties, each in need of assistance from the other two to get what they’re after: consumers without the time, access or inclination to cook or get to restaurants themselves; restaurants who make their menus available for browsing and ordering on DoorDash’s mobile app and website, and freelance delivery people who, armed with the app, are directed to restaurants and hungry customers.
To hear Sukumaran tell it, the model isn’t just funneling business to small restaurants; it’s also shifting how people pay.
Hungry for Treats and Tracking
The company’s target market is millennial consumers, who expect to know where their meals are at each step, and exactly when they will arrive at their doors – as well as whether they have enough time to squeeze in a quick shower while they wait, Sukumaran said.
But this isn’t the only need to be met. The demographic also expects an abundance of options and information to be readily at their fingertips. Calling up a single restaurant, asking about menu items or taking the extra time to place a live voice order is not what these consumers crave. Sukumaran said they want information displayed in a way that presents a wide array of restaurants to choose from, along with reviews, photos and quick ordering.
Most of the company’s customers are either too busy to cook or are without easy access to transportation or restaurants. Its secondary audience is comprised of customers who order to fill a surprise need — for instance, feeding friends who drop by unexpectedly, Sukumaran suggested.
The company doesn’t just cater to consumers’ needs, but also to the needs of the businesses it works with: DoorDash aims to help restaurants that may not be set up to supply their own fleet of delivery drivers or have the same market presence as larger competitors. And, for those who opt to pick up a few delivery gigs, the app provides a mechanism for earning a side income, complete with built-in alerts about when exactly to arrive.
“The way to think about local commerce,” Sukumaran said, “is if you had a way to … efficiently solve the distance problem in an urban context, then you can really power local commerce in a pretty dramatic way.”
Purchasing from a City-Sized Menu
The order-ahead service isn’t just making meals more mobile, however; it’s also giving mobile payments more mileage.
No matter how much a consumer might like her mobile wallet, if most of the merchants she orders from won’t accept it as a payment method, it’s only a matter of time before she also gives up on the idea. This inconsistent acceptance, Sukumaran said, is a major challenge to the broader adoption of Apple Pay and similar methods.
Under the company’s model, it doesn’t matter what payments a given restaurant takes so long as the consumers’ gateway – in this case, DoorDash’s order-ahead platform – accepts them. In this way, the company acts like a city-sized menu, offering access to all participating venues’ items via one ordering interface.
As a result, Sukumaran said the rate of Apple Pay penetration with the service is four to five times higher than the industry average. About 30 percent of the company’s iOS consumers use it for their transactions, in part because the service can save a customer’s preferred payment, making this method easy to reuse.
“We’ve found that there is a shift happening from traditional payment instruments, like credit cards, to things like Apple Pay as well,” he said.
Machine Learning Helps Keep Meals Hot
Being the interface for an array of restaurants is not all fun and games.
It takes serious analysis to ensure that a meal ordered from a local mom-and-pop establishment arrives with the same on-time consistency as a bite from a fast food chain, Sukumaran said. And so, the company works with each restaurant to try to create a smooth order fulfillment system blended with that venue’s operating model – which, for the online startup, could even mean sending orders by fax.
To instill reliable expectations in consumers, as well as to ensure that the delivery person shows up at the eatery right on time to pick up meals while they’re piping hot, DoorDash depends on machine learning. Detailed analysis of a restaurant’s fulfillment time track record, paired with level of demand trends, can help the service predict how long it will take that venue to make a particular item at any time of day – for instance, revealing that ordering dessert from the deli at 11:00 a.m. might require a bit of a wait.
At its heart, Sukumaran said, the company is about logistics, not food, which is evident in its growth plans.
DoorDash forayed into alcohol delivery in 2016, and is exploring prepared meals and flower delivery. Sukumaran revealed the company is even looking into delivering supplies to restaurants, noting that the service can support almost any kind of delivery, as long as it achieves and maintains an efficient network of vehicles, cyclists and people to rally to the task.
While the company works on all that, it will continue touting the message that if customers want to get a bite to eat as soon as possible or businesses need an extra helping of revenue, fast, then mobile order-ahead services might just be the secret ingredient to success.
To get the latest headlines and see rankings of 67 players in the Mobile Order-Ahead space, click the button below.
About the Tracker
The PYMNTS Mobile Order-Ahead Tracker™ serves as a monthly framework for the space, providing coverage of the most recent news and trends, along with a provider directory highlighting the key players contributing across the segments that comprise the mobile order-ahead ecosystem.