It’s So Obvious, Why Didn’t I See It? Visual Business Intelligence
I recently attended a three-day Visual Business Intelligence Workshop presented by Stephen Few. I was happy to hear Stephen cover some of the very basics that so often go overlooked. The workshop started with the basics and progressively built on them. If you are interested in data visualization, I would recommend checking out his website and/or attending one of his workshops.
I often speak with my Business Analysts about the importance of the visual representation of our data. If you provide sloppy or hard-to-read information, the end user could get lost or distracted from what you are trying to show them. Just think about it: if you didn’t comb your hair every day, people may be distracted when talking with you. All kidding aside, when presenting data, you want your point to stand out – or at a minimum your viewer’s attention to be on the data, not on the format.
I want to show an example of a table and a simple chart with issues that may seem obvious, but I guarantee if you look at some of the reports on your desk, you will find at least one of these commonly overlooked issues.
- Useless Information – Don’t show decimals if they are not needed
- Poor Alignment – Make sure alignment makes sense
- Don’t center numbers, always right justify
- Don’t center data titles
- Why are the headers left justified?
- Difficult to Read – No commas used when the number exceeds a thousand
- Misguided Attention – Bold titles draw your attention to the titles, not the data
Another Bad Example:
- Poor Labeling – No title or context to what we are looking at; legend on the side is wasting horizontal space
- Inappropriate Organization – Use of a bar chart when a line chart is more appropriate – we are trying to show a time series trend, not compare cases to hours each quarter; poor use of color – why is hours red? Is something wrong?
- Unclear Formatting – Scale is difficult to read, with two decimal places that clearly don’t need to be shown; no use of commas; no real need for gridlines or a border around the chart
- Scale – Bar charts need to start at zero, the difference in the size of the bars is misleading
If you are paying attention to the data, what it’s really telling you is that complexity is increasing (driving up hours per case) even though call volume is declining. The question we need to be asking: is this a good thing, or do we have a problem? A topic for another post!
These are simple little things you can do to improve the presentation of your information. I picked up some great pointers from Stephen Few and really enjoyed the workshop. He did a great job showing how obvious this is, yet we seem to overlook it. I clearly didn’t do his three-day workshop justice with my simple examples, but I wanted to share some easy pointers with you. Check out your reports, I bet you find some of these commonly overlooked items.