Case Study: Salesforce Experience Cloud Portal

Self Service Portal for Insurance Agents

Built on Salesforce Communities (now called as Experience Cloud) platform that offers a one stop shop for knowledge, case tracking, discussion forums and chat

Duration 6 months

Client: Enterprise Contact Center (ECC), State Farm

Role

Experience Architect

Collaborators

UX Researcher, Product Managers, Technology Leads, Developers

Tools

Sketch/Invision, Freehand, PowerPoint

Problem Statement

Agents use several platforms to find answers or get help, but those tools are scattered and hard to use. As a result, agents often end up calling the contact center—even when they could solve the issue themselves.

This not only increases call volume and costs, but it also slows agents down.

Business Goal

ECC wanted to build a self-service portal to reduce calls and cut support costs—without adding frustration for agents.

UX Opportunity

This presented a clear opportunity to design a centralized self-service portal that would allow agents to easily search for information, access frequently asked questions, and resolve common issues independently—without needing to contact support team.

Discovery

Agent Journey Map

To better understand the problem, I started by looking into how agents and their team members currently find help and support. I wanted to see what their day-to-day experience is like when they need assistance.

We teamed up with a business expert who had actually worked as an agent in the past. Her real-world experience helped us see things from the agents’ point of view and added a lot of helpful insights.

By mapping out the steps agents take to get help, we were able to clearly show the challenges they face. This helped the product team see the bigger picture. It led to full support for a deeper research effort—not just to lower costs, but to create a better, more thoughtful solution that truly works for the agents.

Current State - High Level

Current State - Detailed

Survey

At first, I thought of running a survey to gather insights. I had even drafted some questions. But with the support of the team, we were able to go further—we secured approval for a full research study and brought in a dedicated researcher to lead user interviews. That decision made all the difference.

Research

This wasn’t just about deflecting calls—it was about understanding the real problems so we could build something that agents would actually want to use.

Measure of Success

We would know we were on the right track if

Call volume to the contact center goes down

Agents tell us the new portal makes their work easier

Agents say they’re less likely to call for help

User Interviews

Audience Groups: State Farm Agents (2), Agent Staff Members (2), Contact Center Representatives (2)

Agents

Agent Staff Members

Contact Center Associates

Research Questions

What tools do agents use daily?
What are their biggest pain points?
What makes them call vs. solve something themselves?
What would a helpful self-service experience look like?

Key Findings

Need

Can’t find answers easily
Need step-by-step guidance
Want confirmation before advising their customers

Frustrations

Long hold times
Disconnected chats
Getting passed around departments
Unclear or conflicting answers

Preferences

Some prefer phone support for clarity
Others like chat for speed and multitasking

What They Want

A central place for information
Reliable, updated content
Simple layout that’s easy to navigate
Input from experienced agents
A tool that saves them time, not adds more steps

Define

Agents switch between 3 main systems:

ABS/Agency Hub
Feedback Exchange
Salesforce Answers

These tools work in silos, and agents often create their own shortcuts to keep up. We did affinity mapping to group feedback into:

Pain points
Needs
Likes
Suggestions

Competitor Analysis

We looked at Geico, Progressive, USAA, and others. While they offer digital help to customers, none had true self-service portals for agents. This gave us ideas for features and things to improve on.

Image: Competitor Analysis

Trend Scrapping

I also explored what top tech companies like Apple, Amazon, and Adobe offer in their self-service platforms vs what State Farm agents might need. We pulled inspiration from their best practices to define what our MVP could realistically support on Salesforce.

Data Synthesis and POV: Fear-Hope-Ideas

To make sense of what we heard in the interviews, I organized insights using a Fear–Hope–Ideas framework. This helped me understand the emotional pain points agents were facing, what we can offer them as a hope with Community cloud, and early ideas to guide solution brainstorming. For example, agents felt overwhelmed by scattered information and long wait times, but hoped for guided support, clearer content, and smart tools like video tutorials or personal bookmarks to make their work easier.

MVP & Future Vision

We used our findings to brainstorm key features. This helped us plan for launch and start dreaming about the future. We mapped out:

Ideal Experience (aspirational journey for Agents) vs MVP Features supported by Salesforce Community Platform

Ideal Features

Knowledge search in one place
Ticket creation via forms or chat
Ticket tracking
Community forums with answers from experienced agents
Notifications for updates and changes
Smart routing for after-hours support

MVP Features

Ticket creation via forms or chat
Ticket tracking
Community forums with answers from experienced agents
Notifications for updates and changes

Design

Storyboards

I created storyboards for common scenarios like:

Disconnected Chat – Make it clear and help them resume
Long Chat Hold Times – Show estimated wait and suggest alternatives
Alternate to Long Holds – Allow call scheduling
Time Zone Mismatch – Route to reps in the same zone

Each storyboard focused on:

Reducing confusion
Saving time
Clear communication

Storyboard Presentation Deck

Information Architecture

Page content thought process

Page Layouts

Wireframes

My Contribution

I started by asking: Is this just about deflecting calls?
When I saw the early developer created functional Salesforce prototype, I realized this was much bigger. As someone who uses support tools myself, I knew that unless we got this right for the agents, it would become just another frustrating platform.

So I:

Proposed and conveyed the team for a full discovery phase
Led the Affinity Mapping, Trend Scraping, POV statements
Created storyboards and helped define the MVP
Sketched out early wireframes
Partnered with the researcher and later handed off to another UX designer when I moved to another priority project

This case study really let me blend research thinking with product design, and apply some of new product design methodologies—and I loved it. At the end of the day, any framework that helps us uncover the right problems and design solutions that truly support users is worth exploring.

Final Thoughts

This wasn’t just about reducing call volume. It was about making agents feel confident and supported in their daily work. The more we listened, the clearer it became: agents don’t need just another tool—they need the right tool, built around their needs.

We still had platform constraints, but with thoughtful design and a solid MVP, we were on our way to delivering a solution that actually helps.

Google Sites

Report abuse