After completing the final exams of our technical term in late December and enjoying a short but well-deserved Christmas break back home in Germany, the new year started with the first sprint of our MMA practicum project. The goal of the project is to apply the theoretical knowledge and skill set obtained in the last six months to a real-world business problem whilst working together with an experienced data analytics team. For my team and me, it meant starting our exciting project at Rogers Communications. The project is about developing a machine learning model that predicts the digital propensity of customers during the end of the promotion. Knowing the project's scope beforehand, the goal of this first sprint was to get a feel for the organization, align the business problem with the team, and deep dive into the extensive data available to us.
The first days were spent onboarding, exploring Roger's fantastic office, and meeting all the people involved in the project. After settling in, we reviewed our remaining concerns about the business problem and gained more insights into how our model would be used after the completion of the project. After doing so, we began exploring the various datasets available to us and assessed how each could be used to solve the problem at hand. This experience was different from what I had previously seen in the classroom. Having datasets with billions of rows and thousands of columns, we quickly realized that the pre-processing of the data would take a bit longer than what we were used to from the projects we worked on previously. However, it was extremely insightful to work with a team of experienced data scientists who helped us navigate this exploration phase. After many hours identifying columns, filtering and cleaning rows, joining tables, and engineering features, we ended the first sprint with a preliminary analytical data view of our dataset, which will be used to train our machine learning model.
As a significant part of the MMA program is to develop the ability to communicate the business problem and the proposed solution effectively, the final deliverable of the first sprint was a detailed project proposal in the form of a one-hour presentation. As a result, we had to ensure we had enough time to clearly articulate the problem and structure our proposal so that people unfamiliar with the subject could understand it. After completing the presentation in front of the host company and Rotman staff, we were relieved that our proposal was well received and that we were on a good path to achieving the objectives set by Rogers and ourselves.
I am excited to continue working on this project when we return in March for our second sprint. But for now, I am looking forward to coming back to Rotman each day and switching my work laptop for school books for the next couple of weeks.
The Master of Management Analytics is designed to give students the advanced data management, analytics, and communication skills needed to become an analytics professional