How to handle a giant list of KPIs as a requirement

design process interview

In this article, I'll explain why it's generally not a good idea to build a dashboard from a list of metrics sent to you by the business.

Santa's list

End users frequently desire too many elements in a single dashboard. This bias towards excessive detail, influenced by a spreadsheet-centric mindset, contradicts the principles of effective Data Visualization.

Can I send you my list of KPIs?

The short answer is "no." I believe we should (almost) never start with a list of KPIs emailed by an end user without any understanding of:

  • The strategy
  • The priorities
  • The problems
  • The users environment
  • The future use of the dashboard
  • The complexity of the data and the business rules

Say "yes" and release the kraken

By agreeing to handle this list of 30 KPIs without questioning it, we risk multiplying it by the number of dimensions to use with each KPI.

We will then create charts based on this and finally combining everything into dashboards.

I understand this perspective may seem simplistic, as we rarely design a dashboard for just one KPI. However, I frequently encounter products that contain 15, 20, or even 30 dashboards...

The trick

In order to do so, I use the method from the visual above. This makes it easier to:

  • Customize the dashboard according to user preferences
  • Gain a better understanding of their challenges
  • Enhance the user experience
  • Reduce cognitive overload
  • Minimize filter overload
  • Avoid data overload

You can also take a look at the Dashboard Interview Guide I've created to help you ask better questions in this situation.

- - -

Whenever you're ready, there are 3 ways I can help you:

1 - Unlock The Dashboard Assessment training course and template to help you develop great dashboards that users love

2 - Get your lifetime access to my gallery and learn The Fundamentals of Data Visualization in 50+ minimalist visuals.

3 - The Dashboard Interview Guide will help you ask better questions before diving into your dashboard development.