Context
Insights were strong, but users were overwhelmed
Our healthcare reputation management platform supported marketing and operations teams responsible for monitoring patient reviews across multiple sites, identifying emerging issues, and using feedback to improve service quality.
While users weren’t analysts by title, their workflows were highly analytical: reviewing large volumes of data, identifying patterns, prioritizing actions, and making decisions based on underlying signals.
We had built a healthcare-specific AI capability that analyzed patient reviews, surfaced trends and operational issues, and generated response recommendations.
Approach
The real problem was prioritization, not intelligence
Through customer interviews, workflow analysis, and usability research, we discovered the issue wasn't the quality of the insights. It was how they were delivered.
We had built a system that produced more information than users could realistically process. Customers were already operating in high-noise environments and did not need another dashboard or layer of data. They needed clarity on three things:
-
what matters
-
what requires action
-
why it matters
The core problem was not intelligence. It was prioritization and presentation.
Execution
Redesigning around decisions instead of data exploration
I partnered closely with a product manager on my team throughout discovery and redesign, helping shape product direction, challenge assumptions, and guide strategy while giving her ownership of key aspects of the work.
Together, we redesigned the experience around decision-making rather than data exploration.
This included:
-
Reframing the product around prioritized actions rather than raw insight generation
-
Expanding notification and alerting options to reduce the need for constant dashboard monitoring
-
Introducing configurable digests and bundled insights to reduce cognitive load
-
Restructuring the dashboard to surface the most relevant actions and trends first
-
Allowing deeper drill-down only when users needed additional context
The guiding principle was simple: users should not have to search for what matters.
Outcomes
Faster action and a meaningful jump in satisfaction
After launch with early-access customers:
-
Time spent in the product decreased by up to 20%
-
NPS increased by more than 15 points
-
Users acted on insights they had previously missed due to information overload
More importantly, the product shifted from being a system for exploring data to a system for driving action. Users spent less time interpreting information and more time responding to issues that affected patient experience and brand reputation.
Reflection
More intelligence doesn't automatically mean more value
This work reinforced a core principle in my approach to product leadership: more intelligence does not automatically translate into more value.
In complex workflows, the role of product is often not to increase the amount of information available, but to reduce the cost of understanding and decision-making.
Great UX is often what turns powerful capability into real-world impact.