We had two simple goals for our first Hackathon: We wanted to fix IT perception and to increase collaboration between different IT groups within Friedkin’s companies. We hoped to change perception by showing what IT can accomplish in a short period of time when creativity is encouraged and all day-to-day bureaucracy is removed. We decided to address collaboration by designing Hackathon teams comprising members of different IT groups, hoping that working intensively on a single task would create new relationships.
We started our journey into the unknown by adjusting Hackathon to our company culture. The main change that we introduced was a two day event (Thursday & Friday) with an option to work over the weekend. The second step was asking our companies’ leadership for current business problems. Once we were given this information, time was spent on all of the administrative aspects of running such an event. To reach our first goal, we set up a big event to showcase the Hackathon results for all business leaders, held on the Monday following the Hackathon.
To be honest, running the Hackathon that was a huge gamble. It’s very hard to predict what the results from a 40-year-old IT group with a more laid-back culture will be in a situation such as this. I still remember my managers and me running between the teams’ rooms to make them happy (or in reality to make sure that they didn’t simply go back to their cubicles). The first positive sign was that as the groups spread to their rooms, different business units sent their representatives to lobby for their problems. The second sign was a request for business knowledge help. All of my doubts disappeared on Monday morning when group by group went on the stage and presented what they had managed to do, and they managed to do a lot! One of the groups was able to produce, in two days, a solution that the business had been desiring for twenty years! Our first (trial) Hackathon became a great success; it was so successful that executives asked if we could host one every six months. And that was the trigger for the second Hackathon.
We didn’t have big expectations for the second Hackathon, maintaining that the first Hackathon success was quite enough for us. Reality, though, had different plans. At the final session, each group presented a working solution (not concepts — working solutions!). The second Hackathon ended up being a huge success, with two new solutions deployed to production.
The first two Hackathons reached the two goals we set for them, but the real significant contribution the Hackathons made to our companies became evident after the second Hackathon. Two months later, we had a big issue with our month-end process, which prevented one of our customers (business unit) from closing their month sales. Short analysis discovered that we had several old month-end reports (10 years old) that consumed a lot of our database resources. After another month of trying to solve this problem in the old-fashioned way without any success, we gathered our programmers and challenged them to address the problem as if it were a Hackathon problem. The result was shocking. After one sprint, we had a new solution based on in-memory database to cache data and dramatically reduce database load. The solution was a great success and enabled our business unit to close their month without any delay.
Surprised by the success of this approach and the pace of development, we looked at Team Foundation Server (TFS), where we log and measure our development, to determine whether we could find any correlation between Hackathon and development time. The result was amazing. We discovered that since our second Hackathon, our development time had decreased by 20%!