Following on my last blog post, I am frequently asked what visual to use to display data. While I often hear this question about dashboards, it also applies to many formats including websites, infographics and PowerPoint slides.
There are many excellent sources that describe visualization best practices as well as recommend what data visualizations to use. Here are three that I have used:
1. Dona Wong’s The Wall Street Journal Guide to Information Graphics. For more details, please see http://donawong.com .
2. Cole Nussbaumer Knaflic’s book called Storytelling with Data. For more details, please see http://www.storytellingwithdata.com/book .
3. The web site from data to viz. Based on the type of data you have, the website recommends different types of visuals. In addition, it provides caveats so you can avoid common mistakes.
I hope you find this helpful the next time you are wondering what visualization to use.
As I mentioned in my last blog post, creation of a dashboard should begin with understanding the user, their needs and their priorities. This will enable you to identify the metrics that matter. The next step is organizing those metrics into a dashboard.
You organize a dashboard on two levels. First, each dashboard page should have a consistent look and feel. For example, the filters or slicers always appear on the top or along the left hand side. Once a user is familiar with the layout of one page, it will be easy for them to navigate subsequent pages.
Second, dashboard pages should be organized from the highest to lowest level. You start with the high level key performance indicators so that users can see the state of their business. It may be sales, customer retention, first call resolution, net promoter score, etc. It could also be trends to indicate whether the metrics are getting better or worse. Whatever matters most to the user should appear on the first dashboard page.
The next layer of dashboards should enable the user to do analysis. For example, revenue has declined. Why might that be? If a user has a dashboard that shows the number of customers paying for each product, he can determine if the number of customers has declined or the mix of products sold has changed.
The lowest level dashboards should enable users to drill down to the customer or prospect level and enable users to extract the data in a meaningful form so that they can take action.
Lastly, you should user test your dashboards to make sure that you have captured the metrics in a way that is intuitive and easy to use.
I was asked recently to provide feedback on a dashboard, particularly the look and the feel of the charts. Instead, of evaluating the layout of the page or the use of particular chart types to display data, I started asking questions:
- Who are the users of the dashboard?
- What do they already know about this topic?
- What are their priorities?
- What do they need to know to make decisions or take action on this data?
Often I am asked to critique a dashboard or another form of data visualization. However, you cannot evaluate data visualizations until you know the use case. A use case is the specific situation in which the data visualization, in this case a dashboard, will be used. Without this knowledge it is difficult to tailor the dashboard to meet the needs of the user or the situation.
Once you know the use case, then the questions become:
- Is this data new or does it offer something new?
- Is the data being reliably captured?
- Is the data timely and accessible?
- How can you bring the data to life with visualization?