Combining Data Science and Visualization to Drive Change

John Cook, Senior Director, Data Science and Reporting, US & Canada SMR, Marriott International (MAR: NASDAQ)

John Cook, Senior Director, Data Science and Reporting, US & Canada SMR, Marriott International (MAR: NASDAQ)

John Cook is an expert in resolving complex problems with a range of analytical tools–SAS, Python, R, Tableau–and communicating solutions and insights in a clear and concise way. Cook is well known in the industry for his dynamic presentation and has spoken at internal and external conferences and in countless internal training videos. Cook has led the development of tools and reports across many institutions, such as the University of Maryland, UMGC, and now at Marriott International.

In light of your experience, what are the trends and challenges you’ve witnessed happening with respect to the data visualization space?

I think the biggest trend over the last few years is the movement from more “static” reporting to more dynamic reporting. I was one of the first adopters of Tableau about 6 or 7 years ago, and one of the biggest challenges we had to overcome was a desire to simply have Excel reporting recreated as a Tableau workbook–mostly static tables, pivot tables, and maybe some conditional formatting–to one where people are much more comfortable seeing and sharing numbers in a visual format.

Once people became comfortable that the right visualization could tell a story without having to add bullet points and callouts, I think our ability (as an analytics and data science team) to push different types of visuals helped a lot, and it set the groundwork for the last few years. As a large corporation, I sometimes joke we communicate via PowerPoint, and I’ve seen a trend in many decks to move towards a more visual reporting format. The impact of a well-chosen visualization with a key data point or two labeled is much more impactful than even the best table.

Obviously, Marriott was hit very hard by the pandemic, and we’re still feeling the impacts.

In 2020, we were asked to do a lot more with fewer folks, and I was part of the team guiding the recovery. We were forced to explore new ways to share data, which allowed us to experiment with novel ways of visualizing the information–combining metrics, showing connections, and aligning different points to give a different view–and those have stuck around as we’ve progressed into later stages of the recovery.

Could you elaborate on some interesting and impactful project/initiatives you oversee?

I can’t go into details on specific projects, but I can share that a lot of what we’re working on involves finding new ways to visualize multiple metrics and highlight the right ones, as well as draw relationships between metrics.

Business is so complicated that it’s rarely one or two KPIs that drive change, so we’re looking at multiple KPIs–sometimes 30-40–and finding ways to bring out those most impactful and use them to help drive change. It’s a combination of data science and using visualization to draw the eye to the right changes.

What are some of the points of discussion that go on in your leadership panel? What strategic points do you go by to steer the company forward?

I think the biggest challenge we have is determining how we can make our data more accessible. What do we need to do to ensure people understand it and find the correct insights quickly? Post-pandemic, we’ve significantly reduced the size of Marriott as a company, and all our associates are being asked to do more with less. Sharing data more widely and in a more intuitive format means our leaders at all levels of the organization get the insights they need to make decisions faster and help them maximize their time.

Can you draw an analogy between your personality traits and hobbies and how they reflect your leadership strategy?

I really love to see what other people can come up with, and I’m a laid-back person in general, so I tend to share a problem with my team and leave them alone to create the answer. I don’t need to see the results of every experiment, but I want to be accessible for guidance and to help clear the path and mentor as appropriate. I have an amazing team of people who excel at coming up with creative solutions to the most difficult problems we give them. It’s a lot better for everyone if I let them do what they do well!

I’m also a professor at the University of Maryland Global Campus, and the joy in watching the team create great products is the same joy and pride I have in watching students learn and overcome difficult challenges. It’s extremely satisfying and makes me proud.

How do you see the evolution of the Data Visualizationarena a few years from now about some of its potential disruptions and transformations?

From a corporate perspective, data visualization and customization is going to become more personalized. As we add more and more data to dashboards, users will be able to create their own “custom views” and pick KPIs from a list and have them visualized in a smart and intuitive way. We have a lot of tools for automation on the back end, but we’re only really scratching the surface of what we could do from a user perspective.