Lead Principal Technical Program Manager

Opening Statement (~60 sec)

I took over a program to migrate 2,114 applications off shared legacy AWS accounts into isolated ones — a broad, board-escalated risk mandate with a hard deadline, adoption stuck below 10% after four years, and no one who could say why it had stalled%%. Instead of publishing a plan in week one, I spent three weeks on pure discovery across 11 divisions and three partner teams, then co-built a provisional, explicitly-labeled-as-provisional scoring framework with our security partner so the program could start moving before every detail was final. Within 10 weeks the program had a funded, jointly-owned plan and a moving first wave; within 16 months, adoption went from under 10% to 100% across all 2,114 applications — 2 months ahead of the deadline.

Situation

1. A migration mandate had been open for four years with adoption stuck below 10% across 2,114 applications and 11 business divisions, none of which reported to me, and no prior attempt had produced a clear diagnosis of why it had stalled. 2. Three competing theories were floating around — fluctuating business-unit priorities, no clear ownership or executive mandate, and a hard tooling gap — and I had no standing partnership with any of the three teams whose cooperation the fix would require.

Task

My task was to diagnose why four years of effort hadn't moved the needle, and give 11 divisions a practical decision-making framework to act on before every detail of that framework was finished.

Action

1. 1.Resisted the pressure to publish a plan in week one, and instead ran three weeks of pure discovery — no proposed solutions — asking every division and partner team the same two questions so answers were directly comparable.
2. 2.Co-built a provisional scoring framework (Risk Sloping) with our security partner using the risk data available so far, explicit with all 11 divisions in writing that it was a starting point, not a finished framework, so they could plan against something concrete without being misled about how settled it was.
3. 3.Stood up only the minimum governance the moment required — a RACI and formal risk registration — instead of a heavier structure, so the program could start moving before every open question was resolved.
4. 4.Drew the line between what to commit to provisionally versus what to wait on by reversibility: committed early to cheap-to-adjust decisions (the sequencing methodology), and waited for solid data before asking any division to sign off on an expensive-to-reverse one (a specific funding allocation).
5. 5.Kept a running, visible list of what was still unknown versus locked, reviewed with partner teams on a standing cadence, so "provisional" had a tracked path to becoming "final."

Result

1. The program had a funded, jointly-owned plan and a moving first wave within 10 weeks of relaunch, despite four prior years of stall. 2. Adoption went from under 10% to 100% across all 2,114 applications within 16 months — 2 months ahead of the 2026 deadline. 3. Because no division had been told the framework was final when it wasn't, later refinements to the scoring model never required renegotiating trust with any of the 11 divisions.

Closing Statement (~60 sec)

The mistake I deliberately avoided was treating ambiguity as something to eliminate before acting — I treated *reducing* it as the first phase of the program itself, with its own discovery period and its own deliverable. That's the approach I'd bring to translating a broad OCI priority — improving technician productivity, or operational visibility — into a structured program: diagnose before deciding, give teams a provisional framework instead of silence, and be explicit about which commitments are cheap to adjust and which ones are worth waiting on solid data for.