Leena AI
Employee Feedback to Actionable Insights
Employee Engage was underperforming high support ticket volume, constant client hand-holding. I led the end-to-end redesign: four survey types, four personas, and the full analytics and action-planning experience.
Role
Lead Product Designer
Timeline
4 Months (Launched in 3 Phases)
team
2 Designers & 1 PM
Scope
Dashboard, RBAC, Surveys
Final Results
72% fewer support tickets
64% higher dashboard adoption
4x more action plans created
30% less time to onboard
What triggered this?
I conducted a company-wide audit across all product offerings and found a recurring pattern in Employee Engage. The product was live across 50+ companies and 100,000+ employees, but the interface was aging, and it showed in two places at once: a support queue raising 234 tickets a month across every part of the flow, and a steady stream of feedback from customers and internal teams.
Setting up a survey, reading a dashboard, getting a report to leadership: none of it was self-serve. The CS team was absorbing the gap in tickets.
That wasn't a bug backlog. It was a broken experience, and that is what made the case to rebuild it instead of patching around it.
The real problem
Mapping ticket volume against the product's four core areas showed a clear pattern.
Area | What was broken |
|---|---|
Survey creation | Backend-only flow, so every launch needed manual CS support |
Reports | Nothing shareable, exportable or scheduled |
Analytics & IA | One flat dashboard for every role, so nobody found what they needed |
Action planning | Zero product support for deciding what to do about the results |
This wasn't a UI polish problem. It was an information architecture and role design problem. HR admins, managers, leadership and employees had different jobs to do, and all four were being pushed through one generic flow.
What research corrected
I went in assuming one well-designed dashboard with the right filters could serve everyone. Interviews showed otherwise.
Persona | Core job | What research showed |
|---|---|---|
HR Admin | Create and manage surveys | Wanted self-serve, not documentation |
Manager | Understand team sentiment | Wanted a narrow team view, not an org-wide dashboard |
Leadership | Org-wide visibility | Wanted exportable data, not another login |
Employee | Complete surveys, feel heard | Wanted something quick |
Each persona wanted a different view of the same data, not one shared view with more toggles. A discovery workshop aligned the team on goals, and an opportunity map prioritized where user pain and business impact overlapped most.
The decision that unlocked everything else
Split "survey" from one generic flow into four types tied to real employee-lifecycle moments: periodic, lifecycle/milestone, onboarding, and exit.
Once that existed, dashboards could scope by role instead of stacking filters, reports could shape themselves around a survey type, and action plans could speak to a theme instead of a raw results dump.
Survey setup
Creation moved to a step-by-step flow: questions, audience, test, launch. The old single-page version broke whenever someone hit a launch blocker, with no way back.
Dashboards
Analytics split by persona. A manager sees their team and leadership sees the org.
Reports
They became exportable as PDF and PPT with scheduled email delivery, so leadership gets updates without logging in.
Action planning
Actions got AI-generated, one-click plans tied to specific themes and accepted at the manager level, closing the loop between hearing feedback and acting on it.
Testing
I usability tested setup, analytics and action planning with HRBPs and managers, and A/B tested a conversational survey format against the original linear one. After launch I tracked the same signals that flagged the problem in the first place: CS tickets, dashboard adoption and action plan creation.
Testing led to a simpler launch checklist, anonymous response states and progress indicators that measurably cut drop-off.
Results (first full quarter post-launch vs. prior quarter)
Metric | Result |
|---|---|
Surveys created without CS team help | 75% (9 of 12) |
CS support tickets | Down 72% (about 280 hours saved) |
Dashboard adoption (HRBPs) | Increased by 64% |
Action plans created for identified employees using Gen AI | Increased by 4X (3% to 12%) |
PDFs exported | 10K+ |
Custom reports generated | 500+ |
Looking back
The four-type split was the real unlock, because everything downstream had an obvious place to attach. If I ran this again, I'd build the employee-facing experience first instead of all four personas in parallel. It's the side people touch most often, and getting it right early would have built momentum faster.
The redesign launched in three phases, and the launch slipped slightly. The employee engagement engineering team was carrying a high volume of ad hoc tasks, and because this was a blank project that couldn't reuse the existing code, it needed more engineering time than that load left free. If I ran it again, I'd align with the engineering team on capacity before committing to dates.
