Varcio FinOps Copilot

Opportunity Queue

The cross-cloud triage surface — every detection lands here as a costed, scored finding with SLA tracking, compliance mapping, and evidence export.

At a glance

Route/findings
GroupOperate
Page permissionfindings

What it is

The cross-cloud triage surface. Every detection from every provider lands here as a costed, scored finding, and this is where a team decides what to act on, what to defer, and what to accept — with SLA tracking, compliance mapping, and evidence export throughout.

Who it is for

FinOps practitioners and platform engineers running the day-to-day optimisation cycle, plus compliance teams who need the evidence trail.

How it works

The waste engine executes the detector registry — 305 active rules across AWS, Azure, GCP, OCI, Kubernetes, and cross-cloud patterns — against live inventory, ingested telemetry, and the normalised ledger. Scans run on a schedule constrained by your plan's minimum interval, and can also be triggered manually.

What every finding carries

Estimated monthly saving

What fixing it is worth.

Confidence score

Between zero and one. How certain the detection is.

Effort score

How much work the fix represents.

Risk assessment

What could go wrong.

Recommended action

What specifically to do.

Composite priority

Savings weighted by confidence, divided by effort.

The remediation class taxonomy

Why this taxonomy saves so much time

Nobody has to investigate whether a fix can be automated — the queue already says so. That is often the slowest step in a manual triage process.

SLA tracking

SLA tracking measures how long findings sit unresolved and surfaces the most overdue by breach duration. A triage narrative digest summarises the queue in prose, and root-cause and what-if analysis are available per finding.

Features

  • Unified cross-cloud finding queue with composite priority ranking
  • Per-finding savings estimate, confidence score, effort score, and risk assessment
  • Remediation class taxonomy — advisory, guided, automatable
  • Measured utilisation evidence rendered as a sparkline with p95, average, minimum, and maximum
  • SLA tracking with an aging monitor and most-overdue ranking
  • Root-cause analysis and what-if impact analysis per finding
  • AI-generated triage narrative summarising the queue
  • Bulk status updates across many findings at once
  • Direct auto-remediation trigger from the finding
  • Compliance summary mapping findings to control frameworks
  • One-click evidence export for audit

How to use it

Sort by composite priority and work top-down

The ranking already accounts for savings, confidence, and effort. Do not re-do that arithmetic by hand.

For guided findings, use the supplied command

Lambda memory rightsizing, for example, carries the exact aws lambda update-function-configuration invocation with the recommended value.

Always check the sparkline before rightsizing

Validate the measured utilisation series against your team's dashboards where the workload is unfamiliar to you.

Bulk-dismiss deliberate classes of finding

So the queue stays credible. A queue full of known-intentional items stops being read.

Monitor the aging panel weekly

A growing SLA breach count means triage capacity is short, not that detection is wrong. These are different problems with different fixes.

Why it matters

Most organisations know they have waste but cannot say where it is, what it is worth, or which item to fix first. This module answers all three continuously and across every cloud at once.

The remediation class taxonomy means nobody wastes time investigating whether a fix can be automated — the queue already says so.

Connects to

On this page