Client
Validity
Industry
AI MarTech
Timeline
2.5
Role
Lead designer
Lead researcher
CHALLENGE
Litmus had no structured IA review in over 10 years.
My study surfaced structural problems that hurt usability, discoverability, and customer satisfaction.
Midway through the project, company priorities shifted to Validity Engage, a new AI-powered email testing platform.
The research had been grounded in Litmus, but its findings carried over. They gave me a clear starting point for Engage's information architecture, navigation, and feature scope.
APPROACH
- Week 1
Product audit
Where does the current structure break down?
- Weeks 2-3
Internal interviews
How do our marketing team members work around the current structure?
- Weeks 2-3
Internal card sort
How do our internal team members group features?
- Week 3
Internal survey
What do customer-facing teams see customers get stuck on?
- Week 4
Lo-fi testing
Can 5 real customers find what they need in the proposed structure?
KEY FINDINGS
Litmus's IA was never designed to scale.
The existing structure couldn't scale new features without compounding existing problems, so every addition made the product harder to navigate.
Users can't predict how navigation works
Users can't tell where they are
Users don't know which way to go
Users don't speak the product's language
GUIDING PRINCIPLES
Evolving the IA
The existing structure couldn't scale new features without compounding existing problems, so every addition made the product harder to navigate.
One consistent nav pattern, used everywhere
Breadcrumbs and active states on every page
A single clear path to every task
Language customers already use
INTRODUCING VALIDITY ENGAGE
Pivoting to a new platform
Company priorities shifted, and the planned changes to Litmus were discontinued. We applied what we'd learned to Validity Engage instead, starting from one question: knowing what we know now, what would we do differently?
NEW
Clear hierarchy
Consistent navigation, header, and breadcrumb structure to orient users
Built to scale
Left navigation accommodates room for complex nav structures
Responsive shell
Page framework stays consistent across all widths and viewports
MENTAL MODEL MISMATCH
Collaboration in context, not a destination
The collaboration options in Litmus sat isolated at the end of a 6-step workflow.
Survey responses and card sort patterns surfaced the need to have these throughout the workflow, rather than a standalone destination.
A
6-step workflow
KEY FINDINGS
Outcomes
When company priorities shifted, I carried the findings into the new platform, establishing a structurally sound IA and outlining a navigation structure built to hold up as new functionality was layered in.
As the platform designer, my focus was the customer experience across the ecosystem. Onboarding and PLG were my next priorities to move people through the product smoothly, alongside further developing the design system to standardize the product as it grew.
My employment ended before release, so I can't speak to performance.