← Back to Case Studies
A multi-stage automation system that transformed a manual, error-prone requirements review process into a self-running compliance engine with executive visibility, achieving two consecutive years without an SLA violation.
Requirements review is a critical gate in the engineering development process. Tracking compliance was entirely manual: a co-chair hand-combing through documents, tabulating missed reviews, and drafting follow-up emails one by one. There was no defined SLA, no automated notifications, and no leadership visibility into which teams were lagging. The process depended entirely on one person's time to function at all.
Built in three stages, each replacing manual effort with automation as confidence in the approach grew.
Python extracted compliance data from source documents; results were reviewed and emails sent by hand. Established the data model and validated the approach before investing further.
Email generation and delivery was automated. Scripts still required manual execution, but the notification burden was removed from human hands.
SQL stored procedures replaced the scripts and ran automatically each Wednesday, fully self-running with no manual intervention required.
Every Wednesday, SQL stored procedures queried all active requirements documents and cross-referenced them against the review status table tracking each team's response: approve, reject, object, or N/A. Each non-compliant team received its own targeted email (not a group notification) listing only their outstanding items. As the SLA window passed without resolution, the system automatically escalated to progressively higher leadership levels with no manual intervention required.
All compliance data was written to tables powering a Power BI dashboard, updated automatically each run. Teams tracked their own progress against the OKR. Directors and VPs used it to identify chronic non-compliance and its downstream impact on review timelines.