Case Study: Data Viz & Food Trucks

With a team of three other master's students, I collaborated on this data analytics and visualization design project over 10 weeks, culminating in March 2023.

Our central idea was to research something to do with the popularity of food trucks and street food culture, but we first needed to research all the data sets we could find on the topic in order to build a meaningful and viable data-driven project.  We saw a pattern of food trucks/street food rising in popularity in the United States since 2014, but we did not know all of the factors that led to this rise.

We realized soon, as we cleaned data sets from Google, Yelp, IBIS, and other data sources, that trying to work with data from the entire U.S. was beyond our scope of work.  We learned, while gathering data, that Florida has more food trucks than any other state, so we decided to focus on Florida for more detailed data collection to develop visualizations and tools to be placed in an emerging idea for a final product.

We then ran a survey on channels only for people living in Florida with a series of questions about food truck preferences and habits, rankings, reviews, and other details, to cross-reference with the data sets from Google and Yelp that we decided were the most applicable for our final visualizations. 

As we used our cleaned data (in Excel spreadsheets imported into Tableau) to try out different kinds of visualizations to use in our usability testing stage, we began to design a tool in Figma where we could plug in the visualizations in an easy-to-navigate and interactive dashboard people could use to find the best food trucks in a city, based on the data sets we had collected, and to be able to filter their searches based on a number of factors beyond location, like favorite cuisine and averaged review scores and rankings.

We found we needed to scope further, so we selected one region of Florida for our initial tool - Tampa Bay (including Tampa, St. Petersburg/Clearwater, and surrounding towns) in Central Florida. We added to our Figma-designed dashboard information about that region in case users might be from out-of-state, and we included some history of food trucks as well.

We linked our Figma user-facing tool to our Tableau data and visualizations so users could comb through much more of the data we'd collected, or just use the Figma tool that drew the top-level data from Tableau into the visualizations shown there. We used Schneiderman's Rules to ensure that the visualizations we included were thorough and the best choices for user experience and navigation.