Case study: KFC fixes its performance issues through harnessing its data

Fast food chain KFC has harnessed its data and analytics to improve performance issues and create more efficiency within its franchises. 

Antonio Correa, senior business intelligence consultant at KFC told Digital Nation that in 2019 the popular food chain had a huge issue with performance.

He said restaurant managers were struggling with their reporting platform and it all boiled down to the fact that KFC didn’t have any parallel processing capabilities.

KFC used Snowflake's data cloud platform to help them solve this issue and once that was taken care of, they used other features from the platform to help resolve other problems.

Correa said they took advantage of Snowflake’s data-sharing capabilities to simplify its processes.

“Before we use to share files or replicate our database on cloud databases. It was costly, it wasn't an easy process.

“With snowflake and the data sharing capabilities, we could share data with our media partners. It allows for better marketing campaigns and better utilisation of our data in data science models,” he said.

KFC franchisees are benefitting from this platform as they can access sales data immediately.

“As soon as they’ve got those sales. they can get the sales data. It made life much easier and they can rely on the information, they can load that anywhere using the connections from Snowflake, and our data scientists can now work on the customer segmentation,” Correa explained.

KFC plans to use Snowflake to continue its growth in 2023, opening 25 stores in the next year and enhancing revenue opportunities with partners and the existing 750 stores across Australia

Correa said KFC are looking at using the machine learning capabilities from Snowflake to help streamline processes in its stores.

“[Machine learning] is something that we are keen to do and it keeps out processing unit in one place.

“We want to move IoT and sensor data to Snowflake as well in a streaming way so we can understand the behaviour of the stores and predict possible issues. [We might] send an alert to a store and say there is an issue in that area, please avoid that because it can cause an accident,” he ended.