Retiring a Decade-Old System
Retiring a Decade-Old System
UJA's legacy system had been the backbone for constituent and gift data for more than ten years, built on a platform whose vendor support was winding down. There was no more patching it: the organization needed a full conversion to a new system, not a cosmetic upgrade, re-mapping how 300+ internal users tracked donors, gifts, and constituent records while keeping fundraising operations running through the transition. Midway through the highest-stakes stretch of the build, the project's PM had to step away for an extended personal leave, with no planned backup in place.
A full system conversion, on a fixed budget and timeline, with no room for the leadership gap that showed up partway through.
- FR100 sat on infrastructure the vendor was actively sunsetting, so there was no fallback if the migration slipped
- 116 backlog items had to be sequenced across 6 UI development waves without disrupting daily gift-processing and constituent work
- Internal SMEs, QA, and data architecture all had to stay aligned on one plan and one timeline
- Midway through the highest-stakes stretch of the build, the project's PM had to step away for an extended personal leave, with no planned backup in place
This wasn't a project I was handed to run, it's one I grew into under real pressure, when the person leading it had to step away at the worst possible time. Taking over mid-flight on a system this central to the organization's fundraising operation, with no notice and no handoff plan, was the actual test. Finishing it, on budget, on the wave schedule, with the people who'd use it every day actually trained on it, is what I'd point to first.