Case Study
AT&T Business Transformation
My Impact
Increased sales and and improved self-service experiences for millions of customers by creating simplified core user journeys
Optimized Salesforce’s largest transformation project by consolidating over 100 legacy AT&T systems down to just 3
Empowered design colleagues by creating project templates for interfaces, interactions, and user flows leveraging the SLDS
The Client: AT&T Business
AT&T is an American multinational telecommunications company that is over 100 years old
The world's third-largest telecommunications company by revenue
The second-largest wireless carrier in the United States
The Project: Total Transformation
Transform internal and external sales website, user journeys, and backend infrastructure
Revolutionize AT&T’s entire customer and sales agent experience
Consolidate over 100 disparate systems and remove redundancy
Over 250 members from both Salesforce and AT&T were dedicated to this transformation
My Role: Lead UX Designer
Timeline: March, 2021 to May, 2022
Focus: Late discovery and early design
I concurrently supported design for multiple teams in a Scaled Agile for Enterprise (SAFe) environment
Work streams:
Product configuration
Proposal creation
Contract creation and esignature
Credit check
Post-order enrichment
The Users and Their Environment
User 1: B2B customers
Setting: Using att.com/business to initiate the sales process
User 2: AT&T B2B sales agents
Setting: Using Salesforce to assist customers during the sales process
Fun Fact
The AT&T Business transformation was Salesforce’s largest transformation project to date.
My Methods
I accomplished these goals by collaborating across departments with Salesforce and AT&T employees using the following methods.
Discover issues by researching users and their current experience, pain points, and workarounds
Define challenges and identify opportunities for improvement
Develop answers to those challenges by collaboratively seeking input across departments
Deliver solutions that leverage Salesforce Lightning Design System, Lightning Web Components, and out-of-the-box solutions
My Contributions
Discover
I identified pain points and opportunities by conducting customer and stakeholder interviews
Define
I translated findings from customer interviews into tangible challenges to be addressed
Develop
Collaboratively gathered input across departments
I mapped user journeys, user task analysis, and interaction diagrams which informed our understanding of users
Deliver
I swiftly delivered out-of-the-box and custom solutions for unique challenges by leveraging Salesforce and UX best practices
I created and iterated on task flows, wireframes, interfaces, high-fidelity screens, and clickable prototypes for desktop and mobile screens
Discover Issues
Not All User Research Methods are Effective for Every Project
Discovering Issues With the Issues Discovered
I Discovered issues by researching users and their current experience, pain points, and workarounds.
When I joined the team, the discovery process was already completed. Or so we thought! However, when reviewing the consulting firm’s user research recommendations with the client, we consistently ran into contradictions between user needs and many of the Salesforce platform’s out-of-the-box solutions.
Salesforce Pro Tip
Choosing out-of-the-box Salesforce over customization ensures functionality and reduces the need for future maintenance.
A new Salesforce release can literally break a custom page.
So, we revisited the recommendations and aligned them to out-of-the-box solutions whenever possible.
Additionally, the user research was conducted more like a focus group than like traditional user research sessions. This resulted in blind spots, silos, and recommendations that lacked a holistic viewpoint.
We didn’t have time to start the discovery phase from the beginning. Instead, we uncovered the feature gaps by conducting sessions with various department heads in attendance, observed pain points, and found solutions that met the requirements of all stakeholders.
One of the key findings, was that users had to log in to archaic legacy systems that were plagued with performance issues like long load times, poor error handling, and frequently unmaintained pages. This significantly slowed the process for sales teams to created and close deals with customers.
Newly Minted Findings
Combining the previous findings with these fresh insights, we were able to define the challenges ahead of us, and chart a course for a suite of features that will improve outcomes for AT&T sales teams and customers alike.
Define Challenges
It’s Great When All the Stars Align
How did I Define Challenges?
I defined challenges by listening to stakeholders and synthesizing their feedback to uncover insights.
I Discovered one of my Superpowers: Listening
Gathering input is easy. Creating solutions that meet the needs of various teams, not quite as easy. I enjoy observing users in their element, especially in meetings where requirements between teams may seem at odds. Through careful listening, I was able to define the challenges of multiple teams, and articulate them succinctly to the larger audience. This set the foundation for productive conversations that eventually led to holistic solutions for AT&T.
Which Challenges?
Some of the main challenges we identified
Customer data needs to be entered multiple times across different systems
Disparate synchronization schedules of different systems causes confusion due to no single source of truth
Duplicate data stored in multiple databases to creates redundancy issues and data errors
Users frequently lost their progress and data when they needed to pause or experienced an unexpected error
Develop Answers
Building Consensus = Building Trust
I developed answers by collaboratively exploring potential solutions to the defined challenges. This led to faster conclusions about our constraints, and which solutions to focus our efforts.
Presenting Concepts for Alignment and Sign Off
With so little time, and so many stakeholders, we bypassed the wire framing process in favor of high-fidelity mockups to present for review and sign off. Normally, I would create rough sketches for general feasibility with engineering before creating wire frames. Wire frames would then be created and presented to validate concepts with stakeholders before high-fidelity mockups are created.
Thanks to the Lightning Design System, creating high-fidelity mockups was almost as easy as drag-and-drop. The Configure Product Offer screen shown below became one of the most used templates for many other screens throughout the sales flow. As you can see, there are a lot of choices that need to offer sub choices for customers to select.
AT&T Product examples include:
Fiber Internet
Wireless Internet
Dedicated Internet
Phone Plans
Hotspots
VOIP Phones
Product Configuration template
Deliver Solutions
Done is Better than Perfect
For version 1, I delivered solutions that met the engineering requirements and timeline by sacrificing consistency in the UI.
Version 1
With numerous design commitments and limited time, I had to act fast to deliver the first version of designs for a handful of features. In fact, the whole team was moving at warp speed to meet specific milestones in the project. When this happens, design debt is inevitable. Fortunately, the Lightning Design System took a lot of the guesswork out of the visual designs. What remained, was translating feature requirements, existing workflows, and insights uncovered in the discovery process into elegant, consistent user journeys for mission-critical tasks. Easy right?
For version one, I created complete set of user experience patterns and interactions that could be used as a template for all the upcoming features across the project. A few small improvements, surrounding navigation placement and general layout were all that remained for version two.
Lightning Design System Fun Fact
I worked worked at Salesforce in 2013, and contributed to the Lightning Design System we all know and love.
Credit Check: Sales Agent View - version 1
Version 2: Addressing Design Debt
Once version 1 was under development, I turned my focus towards the long-term strategy of the project. Since we had a holistic starting point, I was able to swiftly address design debt from version 1 caused by technical constraints and project timelines.
Standardized all progress indicator navigation by making them vertical, and positioning them left of the page content
Improved screen and form layout by making compact, scannable, and auto-populated with whenever possible
Updated content copy to align with user tasks rather than database requirements
Created consistency for future designers on the project by designing a global design template
These refinements created alignment across features and also streamlined the development process for future designs and designers.
Credit Check: Sales Agent View - version 2
Self-Service for Customers
With versions 1 and 2 complete and under development, I then focused my attention on providing a way for AT&T Business customers to inquire and purchase products and services on their own. Taking much of the learnings from the Sales Agent flows, I repurposed the template to fit the needs of the AT&T Business customer.
Created a self-service version of the sales flow for external customers
Provided a way for customers to initiate the sales process
Provided a new revenue channel for AT&T
Credit Check: Customer View
Learnings
User research can’t be done in silos
For this project, I learned a number of things about how to better run a large-scale project. For starters,
The right tool for the job Use the right user research methods to uncover insights from your users
Trust, but verify If research has been conducted by an outside team or agency, ensure that the findings are still meaningful before proceeding
Be inclusive If you must revisit user research, gather input from as many sources, stakeholders, and team members as possible
Collaborate and listen Cross-department client discussions in the presence of UX practitioners can save loads of time and frustration
Thankfully, we had frequent design reviews to ensure feasibility and meeting all the requirements. This allowed us to reassess the designs in a more holistic way, and to quickly iterate with key stakeholders to ensure regular alignment on the project.
What Would I do Differently?
Collaborate with Stakeholders Early and Often
One thing I would do differently, would be to conduct our own user research. We had been given user research findings from a consulting agency which had significant deficiencies in their understanding of user needs and the constraints of leveraging Salesforce. We didn’t have to conduct discovery from scratch. But we did spend an unnecessary amount of time at the start of the project proposing solutions based on a flawed understanding of the client. Once we realized that the recommendations we were provided did not fully align to the client’s needs, we were able to have blunt conversations about their needs, and make substantial progress towards a solution.
I would definitely push for these kinds of meetings earlier in the process on future projects. We more easily uncovered the converging and diverging needs of client teams, further clarified requirements, and built trust and rapport in ways previously not possible.
These large-scale projects are often challenging for a number of reasons. I’m so fortunate and thankful that everyone on the client side were willing to have frank and bold discussions about their needs in the presence of non-AT&T employees. This ultimately provided me with the right information to provide best practice user experience solutions for the project and be a part of the project’s overwhelming success.