Copilot Studio, Power Automate, Dataverse, Teams
This project demonstrates how I design end-to-end business workflows using Copilot Studio, Power Automate, and Dataverse, with Microsoft Teams approvals, reusable AI summarization, SLA watcher logic, and strict governance controls built into the lifecycle.
Copilot Studio
Automate
Dataverse
Teams
Azure OpenAI
Entra ID
Power Apps
Security Roles
Manual access requests are often a black hole for IT and Operations. Requests submitted via chat or email lack standardized intake data, making them difficult to process. Approvals require manual chasing, resulting in slow turnaround times, while the absence of a centralized system of record leaves the organization with poor auditability and a weak security governance posture.
I architected an end-to-end access request and approval lifecycle utilizing Dataverse as the central system of record. The solution features Copilot-led conversational intake, automated Teams approval routing, and a reusable AI-summarization child flow. To ensure operational resilience, I engineered an autonomous SLA watcher to chase overdue approvals, backed by comprehensive governance controls and a complete audit trail.
Innovative
Impactful
Customer-Centric
Sustainable
Reliable
Tech-Driven
The architecture relies on a decoupled design where Copilot Studio handles user intake, but Dataverse acts as the absolute system of record. Power Automate serves as the orchestration layer, utilizing a reusable BYOM (Bring Your Own Model) tool configured via environment variables to generate dynamic fallback summaries for approvers.
A dedicated SLA reminder watcher operates as a scheduled flow, autonomously tracking escalation metrics and preventing bottlenecks without manual intervention.
Conversational request intake via Copilot Studio
Reusable AI-generated case summaries (BYOM)
Interactive Teams Adaptive Card approval experience
Dataverse-based lifecycle and status tracking
Autonomous SLA reminders and escalation logic
End-to-end request to fulfillment workflow
Role-Based Access Control (RBAC) implementation
Draft-only editing guardrails and field-level security
Idempotency flags to prevent duplicate processing
Correlation IDs for cross-system tracking
Environment variable-driven configuration
Comprehensive error logging and audit-friendly status tracking
40%
fewer routine email enquiries (opening times, toilets, rules)
65%
increase in “What’s On” page visits
52%
faster content updates by volunteers
>90
mobile PageSpeed score
Building this lifecycle reinforced that a successful automation is only as strong as its governance. While the SLA watcher and correlation IDs provided a highly resilient backend, my next iterations will focus on refining the front-end user experience. This includes developing richer manager context within the Adaptive Cards and expanding the AI-assisted triage capabilities to classify complex, multi-system access requests before they reach a human approver.
