Gabe Orlowitz
- 4 min read

Dashboards Reimagined - A Case Study

Consolidating more than 60 outdated, disparate, confusing interfaces into 3 clear, consistent, and actionable dashboards.

Business objectives:

Increase customer satisfaction
Decrease customer churn
Reinstate ourselves as the industry leader in ratings and reviews

Project length:

11 months

Teams involved:

Product management
User experience (my role)
Visual design
Front & back-end engineering
Technical writing
Client success

Problem

Since its inception in 2005, Bazaarvoice built numerous products to address the same use cases, but executed differently each time. This resulted in client confusion, an increased need for support and services, and significant customer churn.

As the global leader in ratings and review and user generated content, Bazaarvoice helps over 11,500 brands and retailers collect, distribute, and display user generated content on e-commerce sites worldwide. With this comes the need to help clients measure and manage their program performance so they ensure the highest return on investment in our software.

Despite being a ratings and reviews company, answering a simple question like “How many reviews are on my site?” became a noble quest for our users.

The net result was bad for clients, and bad for the business.

Some common pain points for internal and external users alike:

• Widespread confusion caused by discrepant data

• Eroded user confidence and understanding of Bazaarvoice data

• Manual effort required by users to get best-practice metrics

• Outdated and complex BV measurement tools

• Extra internal efforts to resolve client questions

• Engineering tickets to address discrepant data

Objective

Consolidate the numerous, disparate, and outdated data experiences so that our clients log into one cohesive, modern portal where they see clear, consistent, and actionable insights to understand and measure their performance.

Philosophy and approach

At the onset of the project, it was important for us to adopt a philosophy to help guide our decision-making throughout the project. Our philosophy was simple - provide the most valuable, widely used, and actionable insights to the most number of clients. We weren't concerned about giving every piece of data to every single client. In fact, that’s what created the problem in the first place. As the adage goes, if everyone’s shouting, you can’t really hear anyone.

Ruthless intention was the name of the game, and my job was to keep this philosophy alive every step of the way.

My role

Specifically, I was in charge of defining how we told data stories in a clear, actionable, and scalable way across all of our dashboards. I aimed to ensure that we built a strong foundation on top of which we could continue to deliver the most valuable insights to clients.

That meant I was in charge of defining and designing which data visualizations would be needed for which stories, and how the framework that would bring them together.

Process

We questioned every single data story we told. We didn’t just take what existed in the old world and re-skin it. We scrutinized it all. Each data story from the old world went through a series of questions, such as:

Do clients use this?
No really, do clients use this?
How often do clients use this?
Why do they use it?
What do they do with it?
Why was it presented like this?
Is it really necessary?

After continuous scrutiny, we were able to whittle 60+ dashboards into just a few screens. Then, we took our new, much cleaner inventory of data stories and began to identify patterns, which led us to more questions.

Which stories were similar?
Which stories are using the wrong visualization?
What’s the best visualization for this story?

After a massive audit of the current state of our dashboards (60+ screens), and a wealth of client feedback from the years past, we started to break down everything we knew into actionable chunks.

Where were clients visiting the most?
Where were they going the least?
What themes emerged from their feedback? What were we hearing from them?
What is our bread and butter?
What can we lead with first?
What’s the correct MVP to launch with?
What isn’t the right thing to launch with?

We asked these questions, and many more, as we honed in on our first dashboards. From there, we took to the drawing board. Many sketches, many collaboration sessions, internal card sorting, and ultimately, user testing.

Sketches, concepts, and dozens of iterations

This is what months of virtual (Covid-times) team collaboration looks like. Thank you, Invision Freehand!

Iteration upon iteration

Here's an example grid layout structures and card anatomy I explored during the design exploration phase.

We conducted 2 rounds of user testing per dashboard, and immediately converted those learnings into followup design iterations.

One noteworthy finding was that users had a hard time making the connection between which data responded to which control. Due to a lack of visual connection between elements, it was hard for users to grasp all of the changes happening on screen. With this information, we consulted our visual design partner and worked with her to make the necessary changes.

The result:

A single, consolidated, flexible yet digestible dashboard, inviting users right in to measure, manage, and take action on their Bazaarvoice programs.

As with any software design, it's never "done." But, I'm pretty proud of what we accomplished, taking a massive problem and turning into a simple solution.

Thankfully, our users, both internal and external alike, are happy too :)

"Showed my client our new Collection dashboard on our call! They were so impressed with the layout and ease of finding information" - Client Success Director, Bazaarvoice

"I love the new collection dashboard! It has helped with even quick questions (reason for high rejection rates or review origin) that required running a full report in the past." Client Success Director, Bazaarvoice

End of portfolio piece.

Product designer