IRIS Advanced Analytics

IRIS is an internal analytics platform at Kimberly-Clark that reimagines how promotional and predictive data is explored. It helps growth management and trade promotion analysts understand not just what the data says, but why it behaves the way it does.

The project began about two years ago and launched a few months later. Since then, it has never really “ended”—IRIS is an always-evolving product, with new views and functionality added continuously as the business grows and its needs change.

I was the only designer on a 10–15 person product team, owning all UX, research, and data visualization design.

The problem

Before IRIS, analysts relied on Power BI and several external analytics tools that together cost over $2 million per year. These tools weren’t customizable, didn’t quite fit the business’s needs, and forced analysts to work around limitations instead of with the data.

Leadership wanted to lean heavily into forecasting and predictive analytics, but the existing tools weren’t built for that kind of strategic, forward-looking work. Analysts often had to make assumptions simply because the tools couldn’t explain what was really happening under the surface.

IRIS was core to both the main promotional strategy teams and the revenue growth management team. The stakes were high: a massive investment, high visibility with leadership before design even began, and real pressure to prove that this could replace multiple expensive systems. Failure would have looked like people asking for the old tools back.

Goals

The goal wasn’t just to build another dashboard. It was to create a platform that could:

Success meant adoption—if analysts didn’t use IRIS as their primary tool, the project failed.

Users

IRIS is built for growth management and trade promotion analysts. These are highly technical users who work with analytics tools every day. They’re comfortable with complexity, but they don’t want unnecessary friction.

For them, success meant being able to quickly find what they needed, trust what they were seeing, and confidently use the tool to guide major business decisions.

Research

Research for IRIS was deep and continuous. I ran interviews and testing with both stakeholders and real users throughout the process, talking with over 30 people across roles.

The biggest surprise wasn’t resistance—it was excitement. People were genuinely eager for a tool that felt built for them instead of forced on them.

One insight shaped everything: users didn’t just want to see data. They wanted to understand why the data said what it did. IRIS needed to be a tool for explanation, not just presentation.

Core modules

IRIS is made up of several major tools, each solving a different predictive problem.

The Scenario Planner helps analysts simulate how pricing and promotions affect product performance over time. Designing it was incredibly difficult. It had to support complex modeling while still making sense at a glance. Fitting everything in without overwhelming users was a constant challenge.

The Demand Planner visualizes product demand across different product groups and markets. This was the first module we built, which meant it set the foundation for navigation that had to scale as new tools were added later.

Netflow shows how product sales move from brand to brand and in and out of the market month by month. There were no good references for this—it didn’t really exist anywhere else. We had to invent patterns, visuals, and interactions from scratch.

Handling complexity

One of the things I care most about is handling complexity without dumbing it down. With IRIS, we focused on showing what users needed most first, and creating clear affordances for finding deeper detail when they wanted it. That kept the interface clean while preserving power.

We standardized navigation patterns and built persistent filtering so users always felt oriented, no matter how complex the data became.

Collaboration and build

IRIS was built by an internal development team. We ran daily UX–dev syncs.

This was one of the hardest parts of the project. I often had to fight for quality—developers would mark work “done” even when it didn’t meet UX standards. I spent hours in QA pushing for fixes, and sometimes hearing there “wasn’t time.” It taught me how to pick battles, advocate for users, and push for quality without burning bridges.

The only major technical constraint was the charting library, which limited some interactions and visuals.

Launch and adoption

IRIS rolled out in stages as new modules became ready. Power users adopted it immediately, and feedback was overwhelmingly positive.

Over time, IRIS replaced:

It saved the company millions of dollars while delivering something far more tailored to user needs.

Impact

IRIS became the core analytics platform for promotional strategy and revenue growth management. It gave teams the ability to make fewer assumptions and more informed decisions.

Our team earned credibility for being able to build truly complex tools. IRIS became the gold standard for predictive dashboards at Kimberly-Clark.

This is a multi-million dollar success story driven by UX and research. It would not have happened without deep user understanding, constant testing, and a designer willing to live inside complexity.

Reflection

IRIS taught me how to work in extremely complex spaces, how to fight for quality, and how to pick my battles. It pushed every part of my skillset—research, data visualization, interaction design, and collaboration.

IRIS belongs in my portfolio because it proves that UX can drive massive business impact. We didn’t just make something nicer—we replaced multiple million-dollar tools with a platform that does exactly what users asked for.