Data Visualization as a Slide Deck
This project is an example of multiple data visualizations telling a story together in the form of a slide deck. These visualizations are originally made by Cole Nussbaumer Knaflic (see this video), and my effort lies in reproducing that presentation in a web-based form to practice (and demonstrate) my skills with these technologies.
Final presentation
Here is an animated GIF showing all the slides.
Sequence
First of all, the power of presenting data visualizations in multiple slides lies in its sequential nature. It enforces the storytelling aspect of your vis. Instead of presenting a single chart and using it to explain everything, you can create a sequence of copies with slight modifications. Then, you tell the story once piece at a time.
Using color to shift focus
One of the most important differences between slides is the use of color to emphasize pieces of visualization and tune others out. So that each slide focuses on a different part of the story. If the current slide is mentioning some information about products C, D, and E, then the “accent” blue color should be given to the lines of those products while other elements take the gray color to make them less demanding of the viewer’s attention.
Explanatory annotations
I want to emphasize the distinction between explanatory and exploratory visualizations. Explanatory means that the visualization communicates a specific finding or recommendation, which is reached after analyzing the data beforehand. On the other hand, exploratory visualizations are not tied to specific information. They only present the data, and whoever sees the visualization is the one that tries to find insights and draw conclusions. To put it shortly, exploratory visualizations are tools for analysis, and explanatory visualizations are tools for communication.
In explanatory visualizations, annotations are usually a good idea to emphasize the insights presented by the charts. Also, they can be used with color to effectively replace a color legend.
Technologies
This is a web-based project, so it relies heavily on JavaScript.
- D3.js does data-related tasks like importing, manipulation, and scaling the data values as pixel values to be used by the rendered elements.
- React renders and organizes UI elements.
- MDX Deck renders React components as a web-based slide deck using Markdown syntax.