Technical Program Manager

Opening Statement (~60 sec)

I want to tell you about a program where I inherited a 4-year organizational stall and turned it around in 16 months — but the hardest part wasn't technical. At a large financial institution, I led the AWS Small Account (Gen3) Migration, moving 2,114 Business Applications out of a handful of massive, shared legacy AWS accounts — some housing up to 246 unrelated applications — into isolated, dedicated small accounts. After four years, adoption sat below 10%, because no single team could compel 11 independent business divisions, each with its own roadmap and budget, to collectively own a risk none of them wanted. I didn't lead with a mandate. I led with three weeks of listening, then rebuilt the program around making an abstract cyber risk personally real to each division, a partnership model that gave Cybersecurity, our CORE infrastructure team, and AWS ProServe a genuine stake instead of a review role, and an $11M shared-investment pitch that turned 'you must migrate' into 'you're not migrating alone.' The result: within 16 months — two months ahead of the 2026 deadline — all 2,114 applications reached 100% adoption, blast radius risk fell 60%+, compliance rose 90%, and the partnership model I built was adopted as the standard for three other transformation programs.

Situation

1. The financial institution had consolidated over 2,100 Business Applications into a handful of large, shared legacy AWS accounts, with Gen1 accounts averaging 77 co-tenanted applications (up to 246) and Gen2 averaging 35 (up to 126). 2. This concentration created three compounding, quantified risks Cybersecurity had escalated to the C-suite: Blast Radius (a single incident could cascade across dozens of unrelated applications), Lateral Movement (a threat actor compromising one BA could 'role shop' or 'security-group shop' to reach unrelated systems in the same account), and Operational Complexity (IP exhaustion, configuration drift, escalating maintenance). 3. A formal migration initiative had been running for nearly four years with adoption of the safer Gen3 'Small Account' model still below 10% — not because divisions disagreed with the goal, but because no one owned it: priorities kept losing to product roadmaps, the existing tool (Pathfinder) required 3–6 months of manual effort per application, and there was no funding or governance structure to absorb that cost. 4. Three root causes explained the stall: fluctuating business-unit priorities, no clear ownership or executive mandate, and a hard tooling gap that made every migration a slow, inconsistent, manual undertaking.

Task

1. I was brought in as the TPM to relaunch this as an enterprise-wide, time-bound program — drive 100% adoption across 11 Lines of Business, migrating all 2,100+ applications by the 2026 deadline, entirely through influence, not authority: none of the 11 divisions reported to me, and I inherited no pre-approved budget or governance foundation. 2. My mandate had two inseparable halves: motivate 11 competing business units — each with its own roadmap, risk tolerance, and engineering leadership — to align on a shared deadline; and build genuine partnerships, not just coordination, with Cybersecurity, CORE Infrastructure, and AWS ProServe so the program had technical and financial backing strong enough to remove the bandwidth objection that had stalled it for four years. 3. Success meant securing executive sponsorship and $11M in cross-divisional funding for a central migration factory and automation tooling, cutting migration time per application by 70%, and converting every unmigrated application into a formally owned, documented risk rather than an invisible delay.

Action

1. 1.Ran three weeks of structured discovery workshops across application teams, divisional TPMs, Cybersecurity, and CORE engineers before proposing anything, asking only two questions — 'What is making this hard?' and 'What would you need to say yes?' — which surfaced that teams weren't resisting the goal, they were resisting absorbing all the migration pain with zero budget, tooling, or governance support.
2. 2.Partnered with Cybersecurity to run formal threat-model assessments across Gen1/Gen2 accounts, then personally translated the findings from security abstractions ('lateral movement', 'blast radius') into plain-language, per-application risk inventories — opening CIO conversations with 'if this application were compromised tomorrow, here are the other 40 applications in the same account also at risk,' which reframed the conversation from 'when must we' to 'how fast can we move.'
3. 3.Built a tailored value proposition per audience from the same underlying data: 70% migration-time reduction and dedicated factory engineers for engineering leads, a per-LOB ROI model showing the $11M investment would avoid an estimated $120M in future costs for finance partners, and organizational-resilience/reputational-risk framing for executives.
4. 4.Built a genuine partnership coalition rather than a review pipeline: co-developed a Risk Sloping prioritization methodology with Cybersecurity that they owned and trusted, worked with CORE to define migration-complexity ratings and three target-state architecture patterns, and co-designed a two-phase acceleration plan with AWS ProServe — Pathfinder automation for S3/RDS/DynamoDB migrations, then a central migration factory with dedicated TPMs and architects aligned to each LOB.
5. 5.Structured the $11M investment as a per-LOB contribution proportional to each division's application count, then ran alignment sessions with each LOB's finance team to negotiate cost allocation — pitching it as 'you are contributing to a shared engine that does the heavy lifting for all of us,' not a tax on their budget.
6. 6.Resolved the core conflict — Cybersecurity's urgency versus application teams' committed roadmaps, the actual driver of the four-year stall — with three levers: risk-based wave sequencing so the highest-risk applications moved first while lower-risk teams kept flexible timelines; the funded migration factory so no team migrated alone; and formal risk registration in FUSE so any deferral became a documented, owned risk instead of an invisible delay.
7. 7.Secured a signed memorandum from our AE to all divisional CIOs making migration a C-level, 18-month non-negotiable directive, then built the delivery infrastructure to execute at speed — a 9-wave migration strategy tracked via ETBs, CI/CD enforcement blocking new components in legacy accounts, CloudHydra self-service provisioning, and CloudRadar dashboards giving every partner the same real-time source of truth.

Result

1. Within 16 months — two months ahead of the 2026 deadline — all 2,114 applications reached 100% adoption of the Gen3 small-account model, up from below 10% after four years. 2. Blast radius risk fell 60%+ (to under 30% within the first 12 months) and lateral movement risk was eliminated within legacy accounts; compliance posture rose 90%. 3. Automation cut migration time per application by 70%, and CI/CD enforcement blocked 5,459 new legacy components, avoiding $21M in future migration costs and 218,360 engineering hours. 4. The $11M shared investment avoided an estimated $120M in projected labor and cloud costs, and ongoing maintenance costs fell ~25%. 5. All 11 business divisions signed onto one program schedule; the Risk Sloping methodology became the enterprise standard, and the migration-factory partnership model was adopted by three other business units — the biggest win wasn't the metrics, it was that 11 divisions went from avoiding a shared problem to co-owning a shared solution.

Closing Statement (~60 sec)

What made this a motivation-and-partnership problem rather than a pure technical migration was that no authority I had could force 11 independent divisions to reprioritize their roadmaps — the unlock only came from making an abstract risk personally real to each leader, and from treating Cybersecurity, CORE, and AWS ProServe as co-owners with a real stake rather than reviewers I had to get past. The pattern I'd bring here is the same one that broke a four-year stall in 16 months: listen before proposing, translate the same data into the currency each partner already cares about, and design the ask so saying yes is easier than saying no — because at enterprise scale, people don't resist change, they resist being asked to absorb its cost alone.