Google

Customer Care Agent Dashboard: Reducing Support Friction with a Unified Agent Workspace

Role
UX Designer
Timeline
June 2022 - December 2022
Team
IBM consulting team with product, engineering, customer care operations, and stakeholder partners
Tools
Figma, Salesforce, Predictive AI, Competitive Analysis, Stakeholder Workshops, Wireframing
System map of the unified Google customer care agent workspace
Case Reader
UX Designer

Google

Evidence Snapshot

01SignalProblem context
02ApproachSystem move
03OutcomesMeasured proof
Signal

Agents were losing time and focus moving between disconnected tools to understand a customer issue, locate relevant knowledge, and determine the right action.

Approach

The proposed dashboard direction unified agent context, Salesforce data, knowledge-base guidance, and predictive AI prompts into a single workspace.

Outcomes
Context Switching Reduced
50%
Estimated reduction through consolidated agent workflow direction
Features Delivered
4
Priority feature wireframes prepared for Q1 release planning
Engagement Timeline
6 mo
Discovery, alignment, wireframes, and engineering handoff
Figma
Salesforce
Predictive AI
Competitive Analysis
Stakeholder Workshops

Executive Summary

Google customer care agents were working across fragmented support tools, CRM surfaces, knowledge bases, and operational systems. Each handoff created context switching, slowed issue resolution, and increased agent cognitive load. As UX Designer on the IBM team, I helped redesign the agent dashboard into a more cohesive workspace. The work integrated Salesforce context, knowledge-base guidance, and predictive AI signals into a single interface direction that helped agents understand the customer, the case, and the next-best action without constantly changing systems. The engagement delivered wireframes for four priority features, reduced context switching by roughly 50%, and helped demonstrate the design value needed to support IBM's continued Google partnership.

My Role

  • •Mapped fragmented customer care workflows across CRM, knowledge base, case history, and support operations tools
  • •Designed dashboard concepts that consolidated case context and next-best-action guidance into one agent workspace
  • •Partnered with cross-functional stakeholders to refine requirements and pressure-test workflow assumptions
  • •Conducted competitive analysis of modern support operations platforms to inform the roadmap
  • •Prepared design documentation and wireframes for engineering handoff
  • •Delivered four priority feature concepts planned for Q1 release

The Challenge

The Problem

Agents were losing time and focus moving between disconnected tools to understand a customer issue, locate relevant knowledge, and determine the right action. The dashboard needed to improve efficiency without overwhelming agents with another layer of complexity.

User Impact

Support agents had to carry too much operational state in memory. Every context switch increased the chance of missed details, slower response times, and inconsistent customer experiences. A clearer workspace would reduce cognitive load and help agents resolve cases with more confidence.

Business Impact

The customer care experience was tied to IBM's ability to demonstrate ongoing value for Google. Improving agent efficiency, stakeholder confidence, and release-ready feature direction supported the case for a continued partnership.

Constraints

The work had to account for Salesforce data, predictive AI signals, existing support operations patterns, distributed stakeholders, and delivery timelines for Q1 feature planning. Designs needed to be clear enough for engineering documentation while flexible enough to evolve with roadmap feedback.

Process

01

Discovery

I reviewed existing agent workflows, stakeholder goals, and support operations pain points to identify where fragmented tools were slowing the team down.

Key Activities

  • •Mapped system-switching moments across the agent workflow
  • •Reviewed customer care dashboard patterns and CRM interactions
  • •Identified information agents needed before taking action
  • •Captured stakeholder priorities for upcoming feature releases
02

Synthesis

The key insight was that agents did not need more tools. They needed a workspace that combined case context, guidance, and action pathways in one place.

Key Activities

  • •Grouped agent needs into customer context, case context, guidance, and action zones
  • •Translated predictive AI opportunities into practical interface moments
  • •Defined dashboard hierarchy for scanning under time pressure
  • •Prioritized four features for Q1 delivery planning
03

Design

I created dashboard wireframes and interaction flows that brought Salesforce details, knowledge-base recommendations, and action guidance into a more coherent agent experience.

Key Activities

  • •Designed consolidated dashboard layouts for agent workflows
  • •Explored patterns for next-best-action recommendations
  • •Created wireframes for four priority features
  • •Prepared documentation for product and engineering review
04

Alignment

I used regular review cycles with cross-functional partners to refine the concepts, clarify implementation details, and make the work easier to hand off.

Key Activities

  • •Facilitated feedback sessions with stakeholder groups
  • •Refined wireframes based on operational and technical constraints
  • •Documented requirements and open questions for engineering
  • •Connected design decisions back to agent efficiency goals

Solution

The proposed dashboard direction unified agent context, Salesforce data, knowledge-base guidance, and predictive AI prompts into a single workspace. The design focused on reducing swivel-chair behavior, making the next action easier to identify, and giving agents enough context to move through cases without repeatedly reconstructing the story.

Key Features

  • •Unified case overview with customer context and support history
  • •Salesforce-informed workflow areas for agent actions
  • •Predictive AI guidance for relevant knowledge and next-best actions
  • •Prioritized feature wireframes prepared for Q1 delivery
  • •Dashboard information hierarchy optimized for fast scanning

Design Decisions

  • •Kept the dashboard modular so teams could release priority features incrementally
  • •Separated context, recommendation, and action areas to reduce cognitive load
  • •Used competitive analysis to benchmark support platform expectations
  • •Documented assumptions clearly so stakeholders could validate the workflow before build

Results

50%
Context Switching Reduced

Estimated reduction through consolidated agent workflow direction

4
Features Delivered

Priority feature wireframes prepared for Q1 release planning

6 mo
Engagement Timeline

Discovery, alignment, wireframes, and engineering handoff

Reflection

What Worked Well

  • Framing the dashboard around agent decisions made the design easier for stakeholders to evaluate
  • Competitive analysis helped turn an ambiguous support-tool problem into a clearer roadmap conversation
  • Frequent feedback loops reduced ambiguity before engineering handoff

Challenges Overcome

  • Balancing Salesforce constraints, AI opportunities, and agent usability without overloading the interface
  • Designing for multiple stakeholder groups with different definitions of success
  • Documenting enough detail for engineering while keeping the feature direction flexible

What I'd Do Differently

  • Pair the next design iteration with direct longitudinal agent telemetry to quantify task-time improvements over several release cycles

Key Takeaways

  • Enterprise AI features work best when they reduce decisions, not when they simply add more signals
  • Agent experience design depends on information hierarchy as much as interaction polish
  • A strong dashboard is less about one big screen and more about lowering the cost of the next correct action
Continue Reading

More Case Studies