Wave Matrix
Simplifying the Matrix Configuration Flow
From confusion to clarity, driving a 16% lift in sales
Scroll to explore ↓
My Role
Sr. Product Designer
Responsibilities
Defined the problem with PM and stakeholders, led research, validated feasibility and priorities early, and contributed to sprint planning.
Duration
6 Weeks
Collaborate
This wasn’t a design-only problem. so I collaborated closely with Product team(1 VP of Product, 1 Project Manager, 2 Junior Designer), Engineering team(1 Technical Lead, 3 Developers), Support team(1 Support Manager & 1 Support engineer), Sales team(1 Sales Manager & 4 Account Executives), and Customer Success team(1 CS Manager) because Matrix directly affected demos, onboarding, and ongoing usage.
About Matrix
Valorx Wave is a Salesforce analytics product used mostly by the people like CPQ specialists, analysts, and forecasting teams. They deal with a lot of structured data every single day. One of the most important features in Wave is called Matrix. You can think of it like a pivot table for Salesforce data. This is where users compare numbers, spot patterns, and make decisions for their business.
Problem
Users were dropping off early while setting up the Matrix, and many left midway and restarted, which usually indicated they were stuck or confused in configuration part.
How it all started
Why We Looked Into This
While we were getting ready for the next Wave release, we started by checking our analytics dashboard.
And one thing stood out immediately which is Around 40% of users were dropping off within the first two or three minutes of setting up Matrix. Another 28% were leaving halfway and restarting — which usually means they were stuck in configuration part of matrix.
Current Matrix Configuration Screen
Observable User Struggles (Initial Assumptions)
At this point, before doing any formal research, we sat together as a design team with stakeholders and simply walked through the screens. Based on that, we formed some early assumptions about what might be going wrong.
Interface felt dense and overwhelming
Lack of clear visual hierarchy
Unclear structure and navigation
The screen felt crowded, with preview and configuration shown together, which may create conflicting focus. It wasn’t clear whether users should focus on configuration or preview first, and the high volume of inputs and live updates may lead to cognitive overload.
User Problems → Business Impact
Matrix’s complexity made it harder for sales to close deals, resulting in an 8–10% drop in license upgrades and upsell opportunities, and longer onboarding cycles
So we started to talk with teams who interact with users every day. So every team felt the pain, just in different ways.
Support team
told us they received 182 Matrix-related tickets in last three months.
Sales team
shared that in 7 out of 12 demos, prospects disengaged during setup and sometimes AEs skipped Matrix entirely.
Customer Success team
added that Matrix onboarding consistently took longer and needed multiple follow‑ups.
So we focused on goal that Grow Matrix revenue by 10% in Q3 by improving sales conversion and expansion through higher license upgrades and upsells.
Research insights
Finding What Was Really Breaking
Following the signals to uncover what was truly causing friction.
The Question That Triggered Research
At this point, the numbers were clear — but the reason behind them wasn’t. So the real question became
Was Matrix actually too technical for users?
Was the product asking users to make configuration choices without clearly showing what would happen next?
We studied both qualitative and quantitative data using the SPEAR framework.
Ticket analysis and internal
interviews
I decided to work with internal teams to gather data on Matrix. By combining support ticket analysis and internal interviews, we aimed to understand exactly where the experience was breaking down and what's going related to matrix in different teams. Here’s what we found:
What the study revealed
Once we synthesized users interview feedback, into codes and themes, and three themes kept coming up again and again.
Key themes
No Confidence & Decision Assurance
Cause–Effect Visibility
Cognitive Load Management
Theme → Behavior Mapping
Low confidence → frequent restarts & external help
Poor cause–effect visibility → trial and error configurations
High cognitive load → slower completion & abandonment
Root Cause Identified
Matrix combined too many decisions at once, provided too little feedback, and offered no clear sense of progress, forcing users to guess outcomes instead of confirming them.
How others reduce configuration friction
I wanted to see how other configuration-heavy products handle this better.
Key Learnings
Insights that stood out out was that they guide users step by step, ask for structure before data, and let users validate their setup before showing results. These patterns directly addressed the same issues we were seeing in Matrix
Concept Sketch's
From insights to ideas
Using above insights, explored multiple layout directions with the design team. Our goal was to clarify the structure, provide feedback only when useful, and build user confidence step by step. We aligned with PMs on this direction before moving into detailed design — the reasoning becomes clear in the UI solution below.
Design
Building confidence through guided setup
Users weren’t sure if they were doing things right, which led to hesitation and drop-offs. To address this, we introduced a guided step-by-step setup (layout, settings, filters), grouping related decisions into clear steps. Progress indicators and check icons showed what was completed and what comes next, making the setup feel simpler and more manageable.
Testing & Validation
what didn’t work.
We tested the new flow with both first‑time and experienced Matrix users. Overall, users moved faster and felt more confident. But testing also showed gaps — especially around understanding which objects controlled which axis, and how filters were applied
Disconnected Object → Table Relationship
Couldn't quickly understand which object influenced which axis and confusion Between X-axis & Y-axis.
Lack of Visibility for Applied Filters
Wanted upfront clarity on how many filters were active.
Inconvenient Filter Management
Switching filter scopes took too many steps.
Final changes
Clear axis mapping
During testing, users struggled to tell which object controlled which axis and often confused the X and Y axes. To fix this, we assigned each object a distinct color and matched it in the table, along with clear axis labels. This made relationships instantly clear and reduced confusion.
Before
After
One-click active filter control
Users lacked visibility into active filters, and switching scopes took too many steps, breaking their flow. We fixed this by showing active filter counts upfront and enabling one-click switching between X and Y filters with a wizard-style setup, keeping the workflow smooth.
Before
After
Impact
In the end, setup time dropped by about 30%, successful configurations increased from 60 to 72%, which helped more
users reach value faster and support dependency went down. More importantly
we saw a 16% improvement in Matrix-related sales, along with fewer support requests and higher successful configurations.
Before
After
