We visualize data for many reasons. Ultimately, we want to know or learn something about what the data describes. There are many ways to visualize data.
Reports help us dive into the details of a dataset. while some visualizations help us explore relationships in data. In this article we are going to discuss the dashboard, the ruler of here and now.
A dashboard represents the current state, a snapshot, of something you have modeled. It could be the entire business, a department, a customer interaction, operational, financial, or a whole host of other valuable concepts to model.
That snapshot might update in near real time or it might be manually updated and nearly static. It also might look back at historical data or represent the current projection of future performance. Regardless, the idea is to give the viewer the best picture of the model as quickly as possible. In order to accomplish this goal, dashboards must be designed for both simplicity and understanding.
Simplicity
After being in the business for years, consciously or not, you have developed an internal filtering mechanism for what you care about. When you begin to build your metrics into a dashboard, it is important to be very critical about what will go into it because the dashboard doesn’t have intelligence to filter automatically like we do.
Here are a few key ways to make sure you’re designing for simplicity:
- Stay on Topic
The dashboard should be trying to tell a story about one specific topic. Whether it is the strategic health of the organization, the operational health of a department, the effectiveness of a campaign, or the efficiency of a production line, know the topic and be careful not to bring in data from adjacent topics. Knowing your audience for each topic can aid in decisions about what is on topic. - Create Your Own Constraints
Put some constraints on the number or size of the metrics. If you are on a desktop, limit the dashboard to one screen size. In general, you want to minimize the amount of interaction that it takes to get the information you need. - Awareness vs Understanding
The difference between recognition and understanding is important. You can always create or go to additional sources of data to understand why a metric is reading. What are the metrics that trigger awareness of an issue? - The Right Visualization
You can represent data on dashboards in many different ways. Use larger numbers for very important discrete information. Use charts for trends but know they can distort information. Your visualizations should tell a story as quickly as possible while still being clear and accurate.
Understanding
Now that we have simplified the dashboard, let’s look at it from another perspective to make sure we still are capturing everything.
Is it Comprehensive?
First, imagine you have no access to any of the people in the organization. How do you know if the business is functioning well? What if a key employee left. Would you still be able to have your finger on the pulse?
Keep in mind, we are still talking about being aware that everything is going well, but it also needs to be comprehensive enough to minimize surprises.
When designing the metrics for a dashboard, the thought process should be centered around the following question: “If I was on a deserted island, with no ability to communicate with anyone in our business and only had access to this dashboard, what would I need to see on it to get the full picture I need to be aware?”
Is it Actionable?
Is it timely and actionable? Every for-profit business could be simplified down to one metric: profit per year. But is that enough to act upon to ensure that metric stays where you want?
For most businesses, analyzing annually is way too late. At the very least, that metric needs to be closer to real time to be actionable. For a professional service company like Kopis, we need to see revenue daily to have enough time to recognize an issue. It could be a project problem with client impact or it could be a utilization problem.
Either way, finding out at the end of the month or year is far too long to be actionable for us.
It’s perfectly normal to iterate through both Simplification and Understanding multiple times before you are done. In fact, our dashboards have evolved months and even years after we first released them.
