ADMS Upgrade — Data SME
Validated future-state schema against five downstream consumers without engaging external contractors.
Problem
An ADMS (Advanced Distribution Management System) upgrade in a regulated utility is a high-stakes data event. The new schema doesn’t just affect the ADMS itself — it propagates downstream to every system that reads from it. At OG&E, that meant five separate consumers (Oracle, SAP HANA, iDashboard, PowerBI, Nighthawk) all needed their data contracts re-validated against the future-state ADMS model before cutover.
The default option was to engage external contractors for the schema validation work. That was the path of least resistance — and the most expensive.
What I Did
Stepped in as the internal data SME and ran the validation in-house:
- Mapped future-state ADMS tables against every downstream consumer’s read patterns.
- Identified breaking changes in schema, data types, and field semantics — specifically the ones that would have silently corrupted reports if not addressed.
- Coordinated remediation across the five downstream system owners.
- Verified end-to-end data flow in the staging environment before cutover.
The work was technical, but the value was financial: keeping the validation in-house meant the contractor scope evaporated.
Results
| Outcome | Value |
|---|---|
| External contractor scope avoided | $49,031 (documented cost avoidance) |
| Downstream system disruption at cutover | Zero |
| Downstream consumers validated | 5 (Oracle, SAP HANA, iDashboard, PowerBI, Nighthawk) |
| Reports broken at go-live | 0 |
Why this matters beyond the dollar number
The $49K is the headline, but the more interesting story is the pattern. Operations-heavy organizations default to engaging external contractors for schema-validation work because it’s specialized, time-bound, and seen as a one-off. An internal data SME with deep operational-data knowledge can almost always do this work better, faster, and with continuity — because they understand both the source schema and every downstream report that depends on it.
This is the kind of work an in-house engineer with cross-system context delivers that an external contractor cannot.