Understanding Recommendations

CostBeacon runs 42 built-in rules against your Azure resources, cost data, and usage metrics. Each rule evaluates specific conditions and generates a recommendation when an optimization opportunity is detected.

Recommendations are deduplicated using a fingerprint hash — the same finding won't create duplicate entries on subsequent scans.

• Idle Resources — Resources with no cost or usage activity that can be safely removed
• Rightsizing — Over-provisioned resources that can be downsized to save money
• Storage Optimization — Storage accounts, disks, and snapshots that can be optimized
• Network Waste — Unused public IPs, NAT gateways, and load balancers
• Commitment Discounts — Resources with stable usage that could benefit from Reserved Instances
• Scheduling Opportunities — Non-production resources running on weekends/off-hours
• Cost Anomalies — Unusual cost spikes requiring investigation
• Governance — Security and operational best practices (public access, backups, diagnostics)

• Risk Level (Low / Medium / High) — How likely the change could cause issues if implemented incorrectly
• Effort Level (Low / Medium / High) — Estimated implementation effort. Low = minutes, Medium = hours, High = days
• Confidence Score (0-100%) — How confident CostBeacon is in this recommendation based on available data
• Priority Score (0-100) — Combined score factoring risk, savings potential, and confidence. Higher is more impactful
• Estimated Savings — Projected monthly cost reduction if the recommendation is implemented

1. Open — Newly detected, awaiting review
2. Accepted — Reviewed and approved for implementation
3. Implemented — Action taken, savings can be recorded
4. Dismissed — Reviewed and intentionally skipped
5. Rejected — Not applicable or incorrect

After implementing a recommendation, you can record the actual realized savings. This feeds into the Savings Leaderboard and helps track your team's cost optimization progress over time.