Creating a Dashboard: The Do’s and Don’ts

March 10, 2015
A good dashboard maps out key performance indicators
(Photo credit: “Navigator” by Thomas Abbs under the Creative Commons License)

I was recently on a business trip and listening to the Digital Analytics Power Hour podcast (what a great title!) about “What is Dashboarding?” and had some thoughts.  I am doing lots of “dashboarding” myself, so the timing was great.  

After listening to the podcast I thought it was important to first share the podcast and provide my feedback.

The Must’s and Should’s of Proper Dashboarding

The short takeaways from the podcast are the following:

  • Dashboards must be hyper-relevant
  • Dashboards must be organized for the consumer of the dashboard
  • Dashboards must be delivered at the appropriate interval
  • Dashboards should be one page (possibly with supporting pages depending on the audience)
  • Dashboards should provide an at a glimpse view of the KPIs (key performance indicators)

There’s nothing in these takeaways I disagree with. However, there are perhaps some other things to take into consideration. Starting with the KPI (Key Performance Indicators) is a great first step.

Why a Middleman Should Not Define KPIs for a Dashboard

The Hyper-relevance of KPIs need to be determined through strategic planning. This planning has to include the person consuming the dashboard. Period.

Through experience and practice, having a middleman identify KPIs for someone else does not work. The cost of building the dashboard increases rapidly. Inevitably you’ll have to redo a whole lot of work. This usually includes:

  • Data extraction
  • Reworking the cleaning and transformation rules
  • (Sometimes) changing the data visualization depending on the severity of the change

Key Performance Indicators are points of indication and are NOT segmented. This is a critical distinction on why hyper relevance is important. If you’re the Marketing Project Manager for display advertising, your KPI is not going to be net revenue. But it may be display-influenced revenue. You’re not looking at a KPI segment here; you are looking at a different KPI.

Ensure Consistent Clean Data with Automation

An important aspect that was not touched on in the podcast was leveraging automation ability. Automation ensures consistent, clean data. It also includes the tools to manage the connected data and the dashboards.

Dashboards should update fast, and are intended for rapid consumption.  This means that significant time and effort should go into building the first iteration. It should be built to be refreshed and to be automated with minimal to no effort.

A Beautiful Dashboard = Seeing Data with Ease

Though having a pretty dashboard to look at is nice, “beautiful” does not mean fun new ways to look at the data. Line charts, bar graphs, and highlighted metrics are perfectly acceptable means of delivering data. There is a reason they are “common”. They do the job and can do it well. We don’t need bubble charts, doughnut gauges, or multi dimensional area charts just because they look fun.

To me, a beautiful dashboard conveys the data you need to see with ease, not because its a fun colorful chart.  Now, I understand branding can be important. But dashboards are not action-driving on their own. To the wrong person, they won’t provide the context to drive action. However, to the right person, they inspire curiosity and answer questions about recent performance. If more context beyond the basics of the KPIs shown in the dashboard are needed, it was designed for the wrong person. When analysis and insights appear on a “dashboard” then it’s not being used properly. This curiosity and performance evaluation can drive the business forward and solve challenges.

Good Dashboards Inspire Curiosity and Reveal Performance

A good dashboard will inspire curiosity and answer questions about recent performance. It can also reflect performance issues that need attention, and identify big wins that need repeating. A good dashboard guides you to do more of what’s working and less of what is not. It may not be the full recipe for success, but it does tell you what recipe to look for.

We’ve established a bit of what goes into a solid dashboard, in the next blog we’ll talk about how to get there.

Earlier Comments on This Post

Chris: (March 19, 2015) Good article. I have never understood why a dashboard has to be one page. It sounds like underestimating the recipient’s ability to absorb information. What do you think, Justin?

Justin Goodman: (March 23, 2015) 

The idea behind the one page is that the dashboard is quick to digest. It has less to do with the ability to digest information then it does effective use of time. Dashboards should give a rapid status, where if something looks out of line, analysis can be done. A report that may be setup to run at the same cadence can provide more depth information if there is regularly a need to go deeper then the dashboard.

The Idea would be that if the dashboard shows a smoking indicator, a report is already ready and available to support where analysis should could start. The reports, if automated may not always be read, but are available as needed. This doesn’t apply if someone is manually creating the reports. In which case, they should likely be created as needed for efficiency.

As recap, dashboards give rapid status, reports give more in depth information around where further analysis and optimizations can be done.

Paul Dumas: (April 4, 2015) I agree 100% with the entire article. So much, that we have shared it with our clients. Well done! Most of the time less is more for the client.