Portfolio Case Study · Progress Residential
Fourteen teams had been left out of the design. The build was already underway.
I stepped into a Salesforce case management program that had stalled mid-flight. Development was already underway, but 14 of the teams who would live in the system had never been part of the design.
What the program delivered
Figures as reported by the program's executive sponsor. They came from a lot of people doing their part well, and I am proud of the part I drove.
Here is where it started.
- A 40 plus page requirements document the build was following
- 14 functional teams on the Case object, each with its own record type
- Most of those teams never went through discovery
- Leadership had been told discovery was complete
The build was not the problem. The build was faithfully following requirements that described only part of the business.
Why the program existed.
The resident experience organization was fragmented and customer-intensive. Operating costs were climbing and customer trust was slipping. Leadership could not see what was happening in the contact center, the right data was not being captured, and most people were measured on activity numbers that had little to do with whether the resident's problem got solved.
The goal was to consolidate 14 separate issue resolution pathways, and the customer-facing functions and technology around them, into a single standardized model in Salesforce for more than 1,400 users.
If development had kept going, the program would have shipped a system that most of the teams using it had never agreed to. That is the kind of go-live that turns into months of rework.
How we got it back on track.
Working alongside our Senior Process Improvement Architect, I ran the discovery that had been skipped, and together we folded what we found back into the build without moving the release date.
Ran the missing discovery, together
Our Senior Process Improvement Architect and I ran every discovery session with all 14 teams. We confirmed current state with the people doing the work, mapped a future state against the program's objectives, and got sign-off from each team's leader.
Showed every team its future state
Each team saw exactly how the system would be configured for them before go-live. No surprises on day one.
Turned findings into buildable work
I wrote all of the user stories and acceptance criteria. We worked through refinement with the development firm and folded new findings into design and development as they surfaced.
Held the date
The timeline was tight and the release date held. New information kept surfacing, so it was constant adjustment and confirmation with the teams.
Ran QA and UAT as a team
I worked with the QA team and wrote the UAT scripts. My process improvement partner and I trained the teams to run UAT and set them up in a sandbox. Every UAT test had to be signed off by all 14 teams, and every one was.
Prepared the teams for launch
I produced the documentation the training team used to build curriculum for all 14 teams.
Put answers on the case page
I set up Salesforce Knowledge with the knowledge management manager, with subject matter expert review before any article went live. Relevant articles were served right on the case page for the care team.
Gave leadership the view it was missing
After go-live, I worked with the functional teams to build reports and dashboards aligned with what leadership needed to see and manage.
Before
14 teams building in a vacuum, and a build following requirements that covered only part of the business.
After
One standardized model, every team's future state confirmed, and every UAT test signed off by all 14 teams.
No one fixes a program like this alone.
It took the executive sponsor, who championed the program. It took the functional team leaders, who gave their time to discovery and sign-off. It took the development firm, which adjusted mid-build, along with the QA team, the training team and the knowledge management manager. And it took a process improvement partner I worked beside from the first discovery session to the last UAT sign-off.
How I used AI
This was the first project where I used AI heavily, to process and analyze discovery findings, draft process maps, write user stories and acceptance criteria, and build out the UAT process. The judgment stayed with the people in the room. AI helped us cover 14 teams' worth of discovery on a timeline that did not move.
What came next.
The work on this program led to the forming of a root cause analysis team at Progress. I was invited to join it, which is when I moved into my Operational Efficiency Analyst role.
Tools and methods:
I finish what stalled, and I do it with the people around me.