Joe Walsh (2Y 2020) shares his experience of the Analytical Consulting Lab, working with real-world data analytics and the lessons he learned along the way.

ACL Series: Bringing Real-World Data Analytics into Focus

This post is part of a series that dives deep into an experiential learning course, one of Kellogg’s unique curriculum offerings. Keep checking Inside Kellogg for more insights from Professor Shapiro and his students!

By Joe Walsh (2Y 2020)

Analytical consulting lab is an experiential learning course taught by Professor Joel Shapiro, where groups of 4-5 Kellogg students work with a company or “client” to solve a real problem using real data.

I’ve always been interested in analytics, and professionally I want to work at the intersection of strategy, technology, and analytics. The experiential nature of this course really stuck out to me as an opportunity to better understand how businesses really think about data and incorporate it into their strategy and operations to provide real value. So, the chance to work with leaders of an analytics team at a large, international company, while collaborating with a group of good friends here at Kellogg, was something I was really excited about.

Transitioning to a virtual experience

After COVID-19 hit the U.S., the entire class moved to Zoom and I was worried that I wouldn’t experience a lot of the deeper, collaborative aspects of the course. Looking back now, while I do think we missed out on some white board and post-it-note brainstorming sessions, overall, the class, the client, and the team all made it work really well. We met with the client team every Wednesday morning for virtual coffee to update and discuss progress, and we’d meet separately over Zoom as a group 2-3 times a week to bounce ideas off of each other and build out our analyses together.

Real-world data analytics

I’ve taken a number of other data analytics courses at Kellogg, and this course was very different. There’s no regular class, so there was a lot more freedom for us to structure the work as a team. We also learned pretty quickly that in the real-world, data analysis is messy, as opposed to most classes where the datasets you use have already been cleaned and tailored to solve the problem at hand. We spent a lot of our time defining and refining our project scope with the client, and then aggregating and cleaning the data to be able to analyze the question we wanted to solve. It really brought into focus for me that while we hear about all of the great uses for analytics, businesses face real challenges in corralling data and using it in meaningful ways.

Lessons I’ll take with me

The first is that the most important part of data problems out in “the wild” is structuring them efficiently to get an answer. One of my favorite Einstein quotes is “If I had an hour to solve a problem, I’d spend 55 minutes thinking about the problem and five minutes thinking about solutions” and I think the same is true here. The actual analytics is the easier part, while thinking through how to best structure the problem and the available data take the most time.

The second lesson for me is that understanding data analytics and how to apply the methods to real-world problems is very valuable, even if you’ll never be a data scientist. I believe the capacity to understand the analysis process and make connections across an organization’s data sources will put me in a better position to drive action from analytics. We presented our findings at the end of the quarter to a host of executives at the client company, and the ability to translate data to insights and communicate it in a way that was easy to understand led to really good conversations and feedback during the final read-out.

Finally, I think the experiential learning process provides ample opportunity for learning on both sides.  We were able to provide actionable insights and recommendations, and we were able to show the client how we might think about similar problems through a different lens. For me, personally, I learned a ton about a new industry and real-world data applications. These learnings came not only from the client team but also from my teammates here at Kellogg.