The past week was my final week at dv01, a Fintech startup where I’ve spent the last two+ years working on building a product that helps investors analyze their U.S. debt-based structured finance portfolios and markets. I wore many hats—product design and strategy, data viz, snack curation and organization.
I asked my colleagues what they needed from me to smooth the transition, and they said, “can you just teach us everything about data viz?”
I’ve written a bit about my data viz thought process here — I use a version of Richard Feynman’s technique of validating new theories, or Feynmanning, to help me come up with good visualizations for unique data relationships.
I’ll repeat myself:
When I am presented with information about how something works, I start with the simplest representation of the core concept and improvise add-ons as complexities emerge. As new information presents itself, I modify my viz to fit the new criteria. I test it to see if it fulfills all requirements as I go — does it still work in all possible realities within the limits of the criteria presented?
Rinse and repeat, until I have a visualization that fits all criteria and as many realities (i.e. real data, ideally) as I am able to test it against.
But what is the simplest representation of the core concept?
As a prolific consumer and maker of data viz, I have a lot of implicit knowledge of data relationships and general data viz best practices. But I’d never really bothered to attempt to make this knowledge explicit.
So I made a flow chart of many relatively easy-to-implement, well documented (i.e. easily googleable) charts. I’m hoping it’s a useful tool for those who are trying to choose the best viz for the data relationship they are trying to show.
Simpler visualizations are closer to the top of the flow chart, more complex frameworks are towards the bottom. By no means is this an exhaustive list of all possible visualizations or data relationships, but I think it’s great place to start.
Here’s a link to downloadable pdf. Please do let me know if you find this useful or if you have any suggestions for how to improve on this. (I’ll be editing and adding to this as I get feedback!)
List of featured visualizations:
- Bar chart & column chart, stacked variations
- Area charts, absolute and 100%
- Line chart
- Deviation bar
- Bullet bar chart
- Slope graph
- Tree map
- Barbell plot
- Density plot
- Box-whisker plot
- Joy plot
- Bubble chart
- Small multiples
- Sankey diagram (aka Alluvial diagram)
- Gantt chart
- Calendar heatmap
- Choropleth map
- Hexbin map
- Bubble map
- Dot distribution map
Thanks to Frank Deutschmann and Chris Stucchio for feedback!