Data & Advanced Analytics

Using data to craft predictive technology solutions for the world’s largest restaurant company

About Us

Data & Advanced Analytics is divided up into two teams: Data Engineering and Data Science. Data empowers every decision we make, and our decisions have great impact. As the world’s largest restaurant company, Yum! operates 50,000 restaurants in over 150 countries and territories, processes 40 million transactions a year, employs 1.5 million team members and franchise associates, and opens a restaurant every eight hours.

By leveraging cutting-edge, open source technologies and cloud platforms, our Data Engineering team is responsible for architecting, developing and operating the platforms – like data lake, data processing framework, data warehousing and analytics, and data science laboratory – which empower our data scientists and advanced analytics use cases. Our solutions are fully built in the cloud and take advantage of cloud-native benefits such as elasticity, agility and serverless compute capabilities.

Our Data Science team partners with every area of the business to fuel results that have real and lasting effects. Through data, we’ll architect solutions for our Digital Commerce team that will recommend food pairings to customers. We work with our Finance team to ascertain if a menu item performs well in one market, it will exceed expectations in other markets or even on a national level. With Operations, we’ve developed demand forecasting, which smartly predicts how much inventory a restaurant should buy, and with Marketing, we’ve created a predictive email marketing model that serves customers the deals they want right when they’re hungry. And, we collaborate with Development to recommend which locations are the most ripe for placing new restaurants. This is all through data and analytics.

Our collective team is uniquely positioned, simultaneously serving KFC, Pizza Hut, Taco Bell and The Habit Burger Grill, as for the first time, Yum! will be able to store and process data in one common, global location, with a shared infrastructure and management team. We’re now a global data hub for Yum!, facilitating the flow of data between our restaurants, drive-thrus, delivery aggregators, customer marketing platforms and regional brand offices. Our next goal is to enable global insights on our sales, products, and customers – all while using cutting-edge technologies.  It’s a great time to be a Yum data engineer or scientist!

Meet the Team

Portrait of Scott Kasper, Director of  Data Engineering

“I consulted for Taco Bell before moving over to the client-side full time. What intrigued me about working for Taco Bell’s parent company Yum! was the opportunity to expand my scope of knowledge in the data space by working directly with data scientists and the chance to make a global impact. ”

Scott Kasper
Director of Data Engineering

“I have worked as an engineer and data scientist for Phillips 66, Wood Mackenzie and HEB, but what convinced me to join Yum! was the opportunity to work on innovative projects alongside a group of talented data scientists with such diverse backgrounds. Plus, working for a startup within a corporation appealed to me, and once I learned more about the company, the phenomenal people culture stood out. They really care about their employees here.”

Sandra Ezidiegwu
Data Scientist

Projects

Supply/Demand Algorithm

We’re rolling out a new supply/demand algorithm, developed by our data scientists, to 4,000+ KFC U.S. restaurants. The algorithm accurately predicts and suggests orders on behalf of the restaurant manager. We are able to forecast what they're going to sell, how much sales they will do, how many transactions they will have and exactly what product they will sell every 15 minutes.

This saves time and reduces food waste and the need for inter-restaurant food transfers. Because the algorithm is complex, it was taking 15-plus hours per day to run in development – leaving little room for error and making experimentation difficult. By engineering a new solution via leveraging the cloud’s capability to elastically scale, we’ve been able to reduce run time by over 90%, ensuring we’ll always finish our forecasts on time and allowing our scientists to iterate much more rapidly.

Predictive Market Mapping

Traditionally, franchisees and the Development team would determine where a new restaurant would be built based on the amount of sales other restaurants generated in that area and the amount of foot traffic it attracted. We’ve since elevated that process. Now — through advanced data and analytics — we can predict to the dollar how much revenue a restaurant will generate and the percent of sales it may take from neighboring restaurants to avoid cannibalization. We do this by tracking the area population, ascertaining its household income and unemployment rate and calculating foot traffic by the hour. From this, we’re able to identify high potential areas and can recommend that the company develop more restaurants in those areas, positively impacting the bottom line overall.