Top Tips for Presenting Data Visualizations

So you've played around with your data, made some key discoveries that are going to revolutionize your company and now need to persuade others to join your enlightened brigade. The key is turning your analysis into an effective and persuasive presentation. Here are some tips for showing your data visualizations to the world:

Planning

  • Decide whether you're going to start with a conclusion and then explain (deductive approach) or build the story through to the conclusion (inductive approach). Think about your visualizations only once you know you want to present them
  • Write out the point of each visualization slide before designing it
  • For each conclusion from your analysis, try several different visualizations before deciding on the best one to display your data
  • If your conclusion is particularly contentious draw it from multiple visualizations building the story as you go, rather than attempting one super slide to explain everything

Visualizing

  • Make sure your choice is highlighting not exaggerating the trend you're showing
  • Love simplicity. We offer loads of advanced visualizations, useful for all sorts of occasions, but the originals done well often work best
  • Make sure all of the data visualizations presented are key to getting your point across
  • Be consistent with your coloring/labeling all the way through
  • Always show comparisons, not absolutes
  • Correlation and causality are not the same (this one slips through too often)!

General principles

  • It is better to be content rich than design rich
  • Don't make your audience work hard to learn your system - stick to systems they know or link it to something they've seen before
  • Get someone else who has never seen the data before to interpret each graph in 10 seconds
  • Know the weakness of your data and of each graph type and be ready to defend your choices, assumptions and conclusions
  • Be able to show why omitted data is not in the presentation
  • Make sure it's worth telling people about - without an interesting story, no amount of data is going to turn heads

Remember, if you are presenting directly from Bime you can navigate between full-screen visualizations in a dashboard using the arrows on the top right of the screen. You can also filter and manipulate the data on the fly, edit the colors or settings or even change visualization type to make a point during your presentation.

Got any more? Disagree with any of these? We'd love to hear in the comments