We spend our workdays replying to a barrage of emails, shifting priorities, and being in countless meetings. No surprise then that most of us don’t have the time to review a 50-page presentation to get a few nuggets of insight. You say to yourself: “All I want is a 1-pager to tell me everything I need to know and what to do about it”. You open the Executive Dashboard that you have inherited, only to find out that it, too, does not give you any useful answers to your questions. Now you have a 50-page deck, which you have no time to read, and a 1-pager which is too general. Sound familiar? In this post we will explore the 3 reasons why this is happening and what you can do about it.
1. Lack of Focus
A dashboard needs to have a single purpose, a question that it is trying to answer. That purpose is typically to report on business objectives as expressed by KPIs that have been set up within your analytics platform. A dashboard should not be a mish mash of unrelated information such as site visits, instances, bounce rate, time on site, page load times, etc. Those are all interesting metrics, but none of those are KPIs. If your business objective is to sell more widgets, then the entire dashboard should be about lead generation. Forcing an analyst to cram everything into one page might save on paper costs, but it will severely limit the ability to explore that data from multiple perspectives. It’s critical that your dashboard has space for insights that come from the analyst. As the person with the most access to raw data, let your analyst explain WHY things are happening; everyone can see WHAT is happening. The analyst and the business/marketing lead should collaborate to add insights, recommendations and business implications where possible. Keep in mind that a dashboard is not a replacement for a full report.
Like reports, dashboards should tell a story. Every manager/executive has their own version of the truth and their own story to tell. A single dashboard is not able to address stakeholders with different and competing priorities within the organization.
All dashboards should focus on answering a single question. Don’t be shy about turning your 1-pager into a 5-pager, as long as it tells a great story that leads to meaningful insights, actionable decisions and real results!
2. Poor Design Choices
We are visual creatures, our brains process information much faster in visual vs. table form. Sometimes a dashboard can’t tell a good story if it simply doesn’t look good and isn’t easy to interpret. The key question to ask yourself is whether your dashboards are delivering information quickly and easily?
Here are some things to avoid:
- 3D Charts – Not to get too technical, but our brains can’t accurately process size differences in 3D objects. It’s a safe bet to avoid using 3D charts in any of your data visualizations. Just forget that option exists.
- Pie Charts – Similarly to 3D charts, a Pie chart should not be used for comparing more than 5 items. Once you go over 5, it becomes challenging to distinguish the relative size difference between the slices.
- Right Chart for the Dataset – It’s important to use an appropriate chart type to express the data you are working with. While there are no rules about how you should best show your data, we have all been conditioned to expect certain data in a certain format. For example, it’s a best practice to use Line Graphs to show a metric change over time. Play around and see what works best for you and the conclusion you’re trying to illustrate.
Graphs and visuals should further clarify and enhance your data, not confuse them.
3. Rigid Structure
One of the biggest challenges with dashboards and 1-pagers is that they are static, you have no ability to dig deeper. The analyst becomes the gatekeeper to the data, leaving others unable to answer their own questions or draw their own conclusions beyond what has been previously set up. Historically, this has been a major challenge with no obvious solution. In the last few years, there has been a proliferation of Data Visualization tools (Tableau, PowerBI, and etc.) released to market that improve the situation: many of them allow the analyst to create an interactive dashboard and publish it to the web. This allows the end user, with some limitations, to have the ability to change date ranges, apply custom segments, and drill down into the data. While the data is still curated by the analyst, the user is more autonomous and empowered to answer questions and draw conclusions: a vastly improved experience.
A good dashboard gives key data and metrics: an overview. You, the reader, decide what is important. A report should be the opposite: it tells you what is important and then backs it up with details, data, and analysis. The two should live in harmony; a dashboard is not a replacement for solid analysis. It is unlikely that you will get insights from a dashboard - that is not its purpose. In an ideal world, a dashboard is a health check; if things are going south, it’s an invitation to investigate. You can improve your ability to make decisions by creating clearly focused, well laid out and interactive dashboards.