Next week I’ll be speaking at the Gilbane conference on knowledge management. I’m really excited to be presenting on how people find information in the workplace. It’s the result of some really interesting research completed last fall that exposed five unique personas based on how each of them finds information related to their day-to-day activities.
One of the greatest challenges is presenting all the information we found in a simple and succinct way. So, I came up with this infograph to try and capture and communicate what we found related to:
- Their level of domain knowledge, tool knowledge, and experience.
- Activities and tools and how much they used each of these to find the information they need and answer their questions.
- How relevant each of the themes we found were to each of them on an individual basis.
My hope is that after some initial explanation, this model serves better than a bunch of different graphs. Here’s a breakdown of how to read the graph.
- The thin grey lines – represent the entire data set for that persona for that item.
- Thicker lines – represent the majority of the data set for that persona for that item.
- Horizontal line – represent the median point for that persona for that item.
- The pink dots at the top – represent the a unique or important data point for that persona for that item, which is most important. It’s not always the highest in the group, but is one that you should pay particular attention to when designing for that persona.
- The brown dots at the bottom – represent the a unique data point for that persona for that item, which is least important to them. It’s not always the lowest in the group, but is one that you should pay particular attention to as downplaying when designing for that persona.
So, what do you think? Oh, and of course if you’re out at the conference, look me up.
4 Comments
This is very similar (identical?) to a box plot isn’t it?
I’d never heard of a box plot before. But it does seem very similar. I actually adapted some models from Tufte’s The Visual Display of Quantitative Information for this. I don’t recall if he referred to the models I adopted for this as box plots or not. But after reading up on them, it does appear to be very similar.
The exact name for these is “Box and Wiskers”, which have been around for dog years (picked up in 1st semester to graduate quantitative methods in 1993).
There really needs to be (or maybe need to find) a history of information graphic elements. The “tag cloud” has been around for well over 15 years, but known as visually weighted variable sets and there was research on best uses in 1993 (or prior). In grad school we had to be very conversant in Tufte’s approaches, but were warned then that his approaches were best for print and he was not a good historian on info graphics (did not provide attribution well). Tufte does open the doors to visual display of data well for those that are not familiar (like grad students doing deep dives in quantitative methods for social discovery).
I couldn’t agree more, Thomas.