Business Intelligence

Businesses struggle to achieve financial targets and performance goals due to mismanaged and misunderstood data, despite having access to Big Data through BI systems.

🔎 The challenge

The restoration company needed more cohesive data visualization due to their 3-part problem

  • overloaded data systems
  • each employee based their work off different data insights
  • unclear actionable next steps for individuals and leadership

🛠️ The action

We built an immersive experience designed with the following:

  • data visualization dashboards
  • Design system and full icon library

🚀 The outcome

99.6% reduction

in reports generated

45 mins

avg time saved each morning generating reports

5% annual growth

projected increase to business valuation due to increase productivity

0 lost reporting

migration successfully captured all the reporting (daily or longer) conducted
👥 Users: Internal Departments
🗓️ Timeframe: July 2024-May 2025
👩 My role: Design Lead
💼 Our Team: 1 designer, 1 data scientist

  The Initial Discovery

A restoration company was struggling with daily operational clarity due to disjointed, duplicated dashboards and unclear data sources. Reports were generated ad hoc, and employees had to sift through multiple data pipelines to determine what needed attention each day.

On top of this, the company was in the midst of a data platform migration and needed a more effective way to organize and understand its data. I then created a training visual to articulate this to the business.

There were several core users that were needing to access the new dashboard.

  • CEO & Private-Equity Owners
  • Head of Operations
  • Permitting Team
  • Regional Manager
  • Store Manager

We defined the problem space by exploring four key areas, each focused on a different research stream.

  Explore Methodology
⚙️ Method
🛠 Work Conducted

💬

Internal Department Interviews





  • 🪨 employees create a new report every time new data is asked to be interpreted, creating overload
  • 📠 the system and method for data interpretation has been in place for over 10 years
  • 📖 data literacy varies by person and their job description

📊

Reporting Analysis



  • 📑 many reports contain the same information, and often users don’t access them simply because they aren’t aware they exist.
  • 🎉 users access about 10 key metrics on a daily basis and then want to toggle date filters and categories

📄

Market Research

  • 👔 having matching layout between departments boosts recognition and adoption

After mapping out user flows and analyzing over 60 daily dashboards and reports, I found that the overall message was to understand the relationship between install, what projects were scheduled, and capacity, what projects the team could actually complete.

Even with a few requirements, we still had to turn ambiguity into a clear, executable project.

🚪

Low Barrier to Entry

The system must be easily understood and access on current devices.

🔭

Foster Curiosity

Users should feel they can explore the page and feel motivated to learn more about the data shown.

🪞

Match Branding

To integrate with the broader product family, the designs needed to mirror what exists now.

🧠

Match Mental Model

Align with users' existing mental models to avoid disrupting their daily data interpretation.

   Focused Ideation

I storyboarded the workflow of approved and declined results to create a clear sense of place.

My ideation process focused on exploring flexible approaches to data visualization and integration. Since this work is confidential, we were able to explore multiple idea spaces as concepts formed around how to migrate data into specific pipelines—ultimately enabling front-end manipulation for visualization.

Piecing together ideas, I create several paper prototypes to illustrate what we could mockup

💡Idea space: create impression of data relevance

Using individual pieces, I thought through the layout and specific visuals needed. Using a paper prototype, I could easily create new ideas on the spot during ideation sessions and move them around the page as I collected feedback.

Using these ideas, I mocked up new visualizations that combined three related variables—scheduling set, bookings needed, and overall capacity—to reveal key insights.

I also created a design system to improve data literacy by matching recognizable icons to new information.

💡Idea space: improve data literacy through icon recognition

Sketching out different mockups, I explored what iconography would work best with the mental model of our users. This would become the icons used in the visuals, side tab names, and other places for faster recognition and uniformity through out the platform.

   Design & Production

After solidifying our design ideas, I used Figma to prototype our initial concepts and organize what data sources we’d need, along with the backend development required to support the team’s front-end vision.

Using mockups, I blended together our ideas into a cohesive, simple design.

I used the company’s design system—primarily blue, yellow, and custom icons—to highlight key headers, interactive elements, and divide sections within the visualizations.

I also used the bento box method to group together areas on the dashboard according to their relevant and data refresh timing.

I also used emojis on the side tabs for faster recognition and because some tabs were getting cut off depending on window size. Each team had their own set of tabs replicating previous report names they were used to analyzing.

Our usability testing on our pilot and wider launch showed how impactful and effective the data visualziation was our migration was.

99.6% reduction

in reports generated and ad hoc reports requested

45 mins

Avg time saved each morning generating reports and task lists

5% annual growth

With clearer data direction, employees became more productive, ultimately contributing to the business’s overall value.

0 loss of reporting

migration successfully captured all the reporting (daily or longer) conducted

To help with adoption, I created training materials breaking down the design to help learn Power BI and where to find familiar reports.

In future iterations, we could enhance customization by pairing these dashboards with NLP and chat capabilities, trained on company knowledge, without deviating from team understanding.