Public and Utilities
Dutch start-up Quby is best known for its smart thermostat Toon, which puts people in charge of their homes by increasing awareness of energy consumption and costs. Acquired in 2015 by Eneco, a Dutch energy company active in the Netherlands, Belgium, Germany, and the United Kingdom, Quby creates smart energy solutions for homes and offices across Europe.
Better Data, Happier Customers
To help reduce the amount of energy unknowingly wasted by countless households Quby needed smart, data-driven solutions. With Xebia’s help and shared knowledge, it was able to develop new, production-ready data products and a data science workflow. Employees also received intensive training. “From the start, our colleagues were enthusiastic. But as the products advanced, even more people from our organization became convinced of the potential of smart applications to reach our organizational goals,” said Stephen Galsworthy, head of data science at Quby.
New Energy Platform
Xebia advised Quby to start by ensuring data quality and to then make that data available to a small group of consenting users via a new energy platform. Because customer privacy was a key consideration, it required a whole new way of working. Crucially, Quby had to change the way it stored its data: from locally (inside the smart thermostat itself) to centrally (in a Hadoop cluster). To improve one of Quby's core products, Waste Checker, Xebia also developed white goods use cases and identified ways to recognize specific appliances, and expanded applications to include gas. To maintain organizational support (vital for success), Xebia regularly presented progress reports.
Creating a Cluster
“Xebia’s very experienced in developing scalable, robust, and tested data products that can be taken into production easily,” explained Galsworthy. Xebia operationalized a Hadoop cluster by installing data science tools and a workflow that let everyone work on the same reusable code. To manage data while minimizing engineering efforts, it switched to Databricks on Amazon Web Services and facilitated a ten-day Data Science Accelerator Program to improve team skills and establish a true data science culture at the company. “By working closely together and building on experience, we were able to achieve our goals significantly faster,” said Galsworthy.