Data Visualization

Asking the Right Questions

Oftentimes, designing for the easy interpretation of data is an exercise in leading stakeholders through a process which examines their goals for the data.  

My Role
UX Lead, Visual Designer.

Deliverables
Fully designed mockups using dummy data.

The Presenting Problem

To its credit, Southern California Edison knows the value of making data driven business decisions.  And while there is a lot of ‘dashboard’ development, it was rare that  developers ever questioned the list of metrics supplied by stakeholders.  No one asked “Why?”

The result?
Data displays that showed the right information, but weren't very easy to understand.  And which took more iterations than necessary to finally get it right.

My solution? Introduce the UX process

I quickly realized that we needed an actual discovery process, even if abbreviated.  

I devloped a "Dashboard Playbook", which I used to help lead stakeholders and users through a process that asks: Who? Why? and How?

My Process: The Dashboard Playbook

Step 1:  Who are the consumers of the data? 

Stealing from the UX process, we needed to essentially understand the personas of the users of this data. 

The way a CEO looks at the data can be quite different from the way a director might.  And a director in IT may look at the data differently than a director in customer experience.

Step 2:  What questions do they need answered to make their business decisions?

Each consumer of the data needs answers to questions which helps them make decisions to run their business.  I find that understanding those questions makes all the difference in how the data is presented. For example, the way one might display the data for "How are we trending on sales from month to month?" is different than simply displaying this month's sales number.

Step 3:  What are the metrics that help answer those questions?

Only until you answer the previous questions does it make sense to start talking about the metrics needed. I created a table that helps the developers organize the data and its sources.

The Results

Using the process described in playbook, I've created many data visualizations, some of which are shown in examples below.

Covid Preparation Dashboard

In the very early days of Covid, I was asked to design a dashboard that our company used to understand the epidemic's impact on our employees, customers, and supply chain on a daily basis, so execs could make timely decisions about resource deployment.

Within two weeks of working closely with developers, we had a working dashboard.

The Survey Dashboard: Understanding the Status of Inspection Surveys

In the wildfire mitigation effort, to help the Inspections team understand how complete their surveys were, I designed an experience comprised of interlinked dashboards and report pages.

Here is shown a good example of using design to show relationships between the different stats and graphs - color coding in this case.

The stakeholders were very happy with the result and abandoned a previous dashboard in favor of this.  Additionally, this design has been shopped around the company as a model.

Mobile App Dashboard

This Mobile App Dashboard was designed to illustrate the health of our mobile apps, based on retention/adoption, technical performance, and user experience.  It is included as an example of visual design in data.

Michael Blaser

User Experience

mike@mikeblaser.com