AI VOICE EXPERIENCE: AIGENT· B2B SAAS · ENTERPRISE
REDUCING AFTER-CALL WORK BY 25 - 40%
Aigent's AI voice tech was powerful; but setting up the "brain" behind it was a bottleneck that made teams give up. I rebuilt the Real-Time Coaches flow and the live-call view so supervisors could launch AI coaching on their own, and cut after-call work 25–40%.
ROLE
Product Designer - UX/UI & Design System
FOCUS
Real-Time Coaches setup flow, live call monitoring, data visualization
COMPANY
Aigent - Enterprise B2B SaaS
SCOPE
Platform redesign + component library from scratch
2 min read
AIGENT PLATFORM - REDESIGNED REAL-TIME COACHES & LIVE CALL VIEW
THE BRIEF
A powerful AI platform held back by a painful setup.
I led the UX/UI redesign of Aigent's platform and rebuilt how its real-time AI 'Coaches' get created; the exact step that was killing adoption.
Aigent's voice tech worked, but setting up the 'brain' behind it - the Real-Time Coaches - was a dense, technical bottleneck. Supervisors couldn't build guidance quickly, so teams lost interest and gave up. The business needed supervisors to deploy new coaching on their own without calling support, a clear step-by-step flow and an instant read on whether their bots were actually working.
AIGENT - REDESIGNED COACH BUILDER VIEW
THE INSIGHT
The tech was powerful, the setup was killing it.
Customer Success feedback and user research pointed at the same culprit: an interface too dense for supervisors to configure. Three themes emerged.
Conditional logic was too confusing
Users struggled over the basic 'If/Then' flow. Building a rule like 'IF a customer says X, AND the call runs over Y minutes, THEN show message Z' felt like a chore, not a setup.
No clear status tracking
Supervisors were blind to what was running. They couldn't tell which coaching scenarios were drafts, which were in testing, and which were live on the floor.
Disconnected feedback
Agent reactions were locked away from the creator workspace, so supervisors found it nearly impossible to iterate on and improve coaching quality.
USER TESTING
Supervisors needed to see the logic, not fill out forms.
Unmoderated tests with real supervisors validated the new flow; and showed they needed to visualize how the logic connected as they built it.
Breaking up the massive setup form
INSIGHT
Supervisors treated the old setup page like one long, overwhelming document; and missed the logical steps needed to make the AI work.
ACTION
Split the long form into a clean, multi-step tabbed layout. Separating 'Triggers' from 'Messages' forced a clear, natural sequence.
Making the live call data scannable
INSIGHT
The legacy call monitor was dense and monochromatic, so supervisors couldn't scan the screen and spot a call going awry.
ACTION
Introduced high-contrast icons for bot sentiment, agent feedback, and live trigger status; so an anomaly is obvious in under a second.
Fixing the status confusion
INSIGHT
A rough draft and a live coaching scenario looked identical, so it was easy to confuse what was actually running.
ACTION
Added a bold, color-coded status bar to the workspace; supervisors now know instantly whether a bot is in testing or live on the floor.
BEFORE / AFTER · REAL-TIME COACHES
BEFORE
AFTER
IMPACT
Clearing the bottleneck moved every floor metric.
Clearing the setup bottleneck let supervisors launch AI coaching instantly; and the floor metrics moved hard.
−25-40%
AFTER-CALL WORK (ACW)
Less manual wrap-up after every call.
−10-20%
AVERAGE HANDLE TIME
Faster average call resolution.
+8-15%
CUSTOMER SATISFACTION
Higher CSAT scores after launch.
−20-30%
MANAGER ESCALATIONS
Fewer calls handed up to managers.
−30-50%
TIME-TO-PROFICIENCY
New agents reach full speed sooner.
The takeaway I carried out: in complex enterprise tools, information architecture is the real bottleneck; the hard part is translating dense business logic into steps that make sense to a human. Building the feature and a component library together forced every component to earn its place and gave the team a foundation to scale on.
MORE CASE STUDIES
