The US lunch industry is in a state of flux with an ever-increasing number of businesses opting for in-office options vice eating in a restaurant to either increase productivity or provide additional work benefits. Traditionally, many companies have spent large amounts of money on office catering, which largely goes to waste. Thus many firms are turning to technology to answer the question, “what should we have for lunch?” As such, firms now specialize in delivering individualized, restaurant-quality food directly to the workplace. These firms take advantage of the many restaurants offering a wide variety of food near urban offices. However, ordering lunch for a large group of coworkers can be a time consuming task, particularly when attempting to ensure everyone gets what they want. Forkable seeks to bridge all of these elements and use machine learning and technology to match workers to restaurants (food) and vice versa.
Summary of Solution
Forkable uses customer inputted preferences, meal feedback ratings, and individually indicated tastes along with machine learning algorithms to learn the food preferences of office workers and then automate the food ordering and delivery process for individually satisfying office lunches. This approach reduces the complexity of ordering lunch for large office co-worker groups by not only predicting what restaurant and food every person in the group will enjoy, but also completing the ordering, pickup, and delivery of those lunches. Forkable also allows for users to meet budget constraints and use human override if required. Additionally, Forkable eliminates the waste associated with catering corporate lunch as well as cutting down on the internal administrative cost of ordering lunch, saving companies money.
Effectiveness and Commercial Premise
Forkable is part of the emerging space that is the intersection of the $9B catering and $200B U.S. lunch markets. In this area, firms typically charge a delivery fee plus an order surcharge to generate revenue. As Forkable is still very much in the startup phase, it remains impossible to directly measure either the potential or effectiveness of Forkable. However, Fooda may provide a closer benchmark by delivering meals from a single restaurant, earning $48M in revenue in 2016. At the top end, Grubhub provides a measure of a large, public company that allows one to achieve the same results with more effort required from the user, generating $683M in revenue in 2017.
Lunch catering is a crowded and diverse competitive landscape. There are low barriers to entry and a wide variety of approaches to catering. There are traditional businesses that have a limited tech presence, but will cook, deliver, and serve the food. There are also a variety of well known tech-enabled businesses, like Grubhub, Postmates, and Uber, which focus just on the delivery of the food from restaurants to customers. Other lesser known office-catering businesses include Fooda and Eat Club. Fooda’s business model involves bringing local restaurants into offices to serve a selection of food directly to the customers. This model is interesting and can operate with low overhead, but a narrow selection from a given restaurant will not necessarily have something that everyone likes. Eat Club has a model that is similar to Forkable in that it allows customers to individually select their preferences. In this way, it is similar to just ordering from a menu. It does not use machine learning to select dishes and restaurants which are best suited to customer preferences in the way that Forkable does.
Proposed Alterations to Increase Value
Forkable currently optimizes meal selection for the customer through the previously described preference-rating feedback loop. However, restaurant selection occurs by using websites known for ratings, such as Zagat, Yelp, and Foursquare, and selecting 10 restaurants in a given area from which to offer meals. As a result, the process could be improved by using machine learning to optimize restaurant selection. Forkable would take the upcoming week’s customers (which they know in advance) and identify the restaurants through machine learning within a predefined radius that provides the highest number of potential matches for the customer set. This process would accomplish two goals. First, it would increase the quality of the service and customer satisfaction as a result. Second, it would decrease costs through both restaurant search-radius optimization (i.e., pick an adaptable area that will minimize delivery costs) and decreasing the cost of customer acquisition/retention. By offering a higher quality service at lower cost, Forkable would increase its customer base, attract more restaurants, and be able to offer more options, all part of the virtuous flywheel associated with indirect network effects. As a result, this improvement would both create value as the emerging market size increases, as well as capture value by removing other close competitors such as Fooda and ZeroCater.
Forkable Wants Companies to Forgo Buffet Style Lunches
Food Tech Startup Marries In-office Catering with Recurring Revenue Model
Thomas DeSouza, Matthew Nadherny, Patrick Rice, Samuel Spletzer