Hello OAC Community!
This is my very first time building a dashboard, and I’m excited to share how I used Oracle Analytics Cloud to transform a standard procurement dataset into a strategic Compliance and Risk Dashboard.
The Dataset & Problem:The source data tracks approximately US$37.8M in organisational spend across core suppliers from 2022 to 2024. While the raw data showed "what" was spent, it hid a critical story regarding Procurement Maturity and Policy Compliance.
The Big Idea: The "Transformation Narrative"My primary goal was to move beyond descriptive "spend by supplier" charts. Instead, I focused on identifying policy leakage and demonstrating risk mitigation over time.
Key Analytical Features:
- The Compliance Timeline: Using a 100% Stacked Bar by quarter, I visualised the journey from a high-risk environment (30% non-negotiated spend in early 2022) to achieving 100% Negotiation Compliance by Q4 2023. This highlights operational improvement rather than just static totals.
- Maverick Spend Detection: I implemented a specific "Maverick Spend" view to isolate suppliers receiving funds without formal agreements. This helps identify "leakage" where the organisation may be overpaying for market-rate items.
- Supplier Risk Concentration: Using Treemaps, I visualised spend density by supplier and item description. This revealed that high-ticket items, such as Laptops (US$355K Unit Price), are now successfully governed under negotiated terms.
- Predictive Foresight (ML): On a separate canvas, I leveraged OAC’s Time Series Forecasting to project future spend, allowing the procurement team to proactively manage upcoming contract renewals based on predicted volume.
Strategic Conclusion: By auditing Agreement Flags against Negotiated Status, this dashboard proves that the procurement team has effectively secured financial interests. I have successfully shifted from unmanaged "Maverick" spending to a 100% governed ecosystem.
I would love to get the community's feedback on:
- How do you typically handle "Maverick Spend" visualisation in your OAC environments?
- Any suggestions for further ML-driven anomaly detection on this type of indirect spend data?
