Dangy is a mobile app that uses AI to assess the user’s taste profile, find users with similar taste profiles, and analyze ingredients in various restaurant dishes to recommend new dishes that the user would love.
What inspired you to start Dangy?
“Cheat Day!” It’s one glorious day each week where our family breaks away from our strict diet to eat anything and everything we want! This is such an important day that the whole family spends the entire week planning out what to eat for Cheat Day. It’s too precious to eat just any unhealthy food. It needs to be perfect and fulfilling, so we could power through another miserable week. So around the dinner table, everyday, the endless debate was: “What’s for cheat day?” We then found that lots of other people had the same problem of not knowing what to eat. So, we set out to fix that problem, and Dangy was born.
Why is this problem important?
Americans love to eat out. 238M eat out at least twice a week, 271M try new restaurants, and 195M use restaurant reviews. Interestingly, 91% of Americans try new restaurants because of friends’ recommendations while only 68% use websites like Yelp and Google for such purposes. This is because review sites only rate restaurants while friends recommend dishes, and one knows whether their friends’ tastes match theirs. Our customer interviews say that people tend to “stick to usual orders unless friends recommend something” and are wary of waiters’ suggestions because people have “different tastes” and “waiters might recommend a more expensive dish for higher tip.” Diners need trusted recommendations for new dining experiences. Dangy is designed to give new restaurant dish recommendations that the customer can be confident they will love.
In addition, interviews with restaurants have shown that they are failing to reach new customer-bases as “[many] only promote through social media, email, and [their] mobile app.” This is because restaurant advertising is expensive and largely ineffective. Platforms like Yelp and Google Ads have a bidding process behind the scenes that drives up the cost-per-click to $40 and $10 respectively. Online reviews of Yelp for Businesses also show that 85% of businesses surveyed rated the service one out of five stars with Forbes magazine saying, “Adwords just doesn’t fit. It’s better to find an alternative that actually gels with your niche.” We’re solving this problem by eliminating the bidding process behind the scenes to ensure that all restaurants have equal opportunity to advertise their dishes. Furthermore, we are making advertising more effective for restaurants because we know the taste profile of our users. So, we will be matching users with advertisements that feature dishes that we know the user will love to increase click-through rates and conversion rates for our restaurants.
Finally, while 32% of restaurants analyze menu reports, they state that they “…only know how many of each dish [they] sell.” “The most useful thing [they] wish [they] knew is why customers don’t like certain dishes.” Dangy’s platform directly collects this information through dish reviews. This will be sold as a subscription service along with other data to help restaurants better understand their customers.
Each time a customer comes to a restaurant, that restaurant has one chance to make a good impression. But restaurants have little ability to ensure that customers eat the dishes that best fit their tastes. By recommending the best dishes for customers at a given restaurant, Dangy will be increasing the lifetime value of each customer for that restaurant so that customers leave with happy bellies and great impressions.
Tell us about how Dangy has grown over time.
Dangy started as a “what-if?” idea midway through 2020. However, Elijah and Jason had never made a full mobile application and had no experience building review platforms. So, Elijah and Jason spent their free time outside of internships and other summer obligations to work on a simpler mobile app called Co-Crit. Like Dangy, Co-Crit crowdsources reviews, and Elijah and Jason used it as an opportunity to learn how to effectively market Dangy, how to engage users to write reviews, and how to build a scalable mobile app. After Co-Crit’s completion, Elijah and Jason began conducting market research, writing code, and developing the business model for Dangy. Since then, Dangy has validated its need through extensive user interviews, built its first AI model which has accurately ranked dishes with 91% accuracy, launched its website, developed a sustainable business plan, and is about to launch its Alpha.
Who is your target market?
As a two-sided marketplace, Dangy brings users and restaurants together. For users, Dangy is useful to everyone. For foodies, Dangy helps you explore new dishes and be adventurous! It helps you share your favorite dishes, learn about hidden gems, and try new cuisines. For those who don’t enjoy trying new foods, Dangy seeks to change your perspective. Specifically, we have found that most people do not enjoy trying new foods because they are afraid of ordering something that they do not like. Dangy solves this problem by matching users with dishes that they would enjoy!
In breaking down our target market further, Dangy will initially target millennials and Gen Z foodies who eat out once every two to three days and are located in Seattle, Washington.
For the restaurants, Dangy will initially target local restaurants in Seattle. Specifically, we are targeting local pubs and grilles that are at “Usually not too busy” or less (according to Google) 80%+ of the time. Pubs and grilles are frequented by millennials and Gen Z foodies, tend to have an extensive list of dishes and drinks, and are prime for special offers that can be promoted through our platform. Plus, they have the greatest profit margin out of any other type of restaurant!
How is B-Lab helping your venture develop?
B-Lab has been tremendously helpful in providing feedback on Dangy’s launch strategy. Specifically, B-Lab lectures and meetings with our mentors provided through the program have pushed us to speak to more users, obtain quantitative data that validates the need for Dangy, and refine our business model. Through B-Lab, we have refined our value proposition to get closer to product-market fit, grown our network of resources that we can turn to for help, and established a stronger understanding of Dangy’s mission and vision.
What is something surprising that has happened thus far?
We were surprised by our user interviews. As we spoke to people of different backgrounds, we found valuable insight into the problems that our users face. Specifically, though seemingly obvious, we found a recurring problem that Dangy can uniquely solve. Many users struggle to decide where to eat when they are with other people:
Person 1: “What should we eat?”
Person 2: “I don’t know…what do you want to eat?
Person 3: “I’m good with anything!”
Person 1: “I don’t know. How’s Italian?”
Person 2: “Eeeh, no let’s get something else.”
Person 1: “Chinese?”
Person 3: “No! Definitely not!”
1 hour later…and you still don’t know where to eat. This situation isn’t new to any of us. Dangy solves this uniquely because our AI model knows exactly what places best fit everyone’s tastes! This led to us to create a “Party-Up!” feature that allows you to find restaurants best suited for everyone in your party and tells you exactly what each person should order.
What’s your favorite homemade dish?
Our parents are fantastic cooks! Our mom makes a killer roll that has tempura shrimp, fish roe, a variety of veggies, and a house-favorite chipotle sauce! Our dad grills up a perfectly seared Galbi (Korean short-rib) that’s sweet, smokey, and melts in your mouth! But our family loves to eat out and explore. Whether it’s going to eat Chicago-Style Deep Dish (yes…better than in Chicago!), crispy peking duck, or Instagrammable bread pudding with white chocolate and raspberry sauce, we’re always thinking about what to eat next!
Anything else you’d like to share?
Yes! We want to invite you to join our Early Access List so that you can be the first to discover restaurant dishes you’ll love! You can also check out our website at dangyapp.com to see what our app looks like and follow us on TikTok, Instagram, and join our Facebook Group!
Jason Whang (right) graduated from Brown University magna cum laude in 2021 with a diploma in Business, Entrepreneurship and Organizations: Tech Management (BEO). Jason is an incoming software engineer at Amazon. He has previous experience as a researcher for Brown’s Artificial Intelligence (AI) Lab, presented at the 14th NY Machine Learning Symposium, worked as a marketing and business intern for three different startups, and was a Teacher’s Assistant (TA) for Brown’s CS16: Introduction to Data Structures and Algorithms course. Jason also has previous experience as the founder for another venture, which won 2nd place for Google Cloud solutions at the 2020 Hack@Brown competition. Jason was also a part of the Brown Men’s Club Basketball Team for four years, a Residential Peer Leader (RPL) for three years, and an active member of Brown’s BEO Department Undergraduate Group (DUG). Jason’s favorite dish is Chicago-Style Double Spinach Deep Dish Pizza!
Elijah Whang (left) is a rising high school senior at Basis San Antonio – Shavano Campus High School. Elijah is currently an AI researcher at the University of Texas at San Antonio and was a finalist in the 2020 Regeneron International Science and Engineering Fair for his project Tracking Red Blood Cells in Real Time with Deep Learning to Predict Mortality. He was a keynote speaker and 1st place winner for the Computer Science and Math category at the Alamo Regional Science and Engineering Fair. He also has gotten a perfect superscore on his ACT (36), taken 21 AP classes, and coded his own personal website. He is a four-year varsity letterman for his high school basketball team, three-year team captain, and is involved in other leadership positions,such as Mu Alpha Theta and National Honor Society. Elijah’s favorite dish is Galbi (Korean Spare Ribs)! Finally, Jason and Elijah are also the founders of Co-Crit. Co-Crit is a mobile app that was featured on 10 news outlets including the Boston Globe.