What is Flux

One governed layer over cost, inventory, utilization, and governance — with the evidence behind every number kept alongside it.

Flux is a cloud intelligence platform for FinOps and cloud operations teams. It continuously collects cost, inventory, utilization, and governance data, keeps the history, and turns it into findings you can act on and defend.

The problem it solves is mundane and expensive: cost lives in one export, inventory in another, utilization in a monitoring tool, and governance in a portal blade. Reconciling them takes days, and by the time anyone finishes, the estate has changed. Four teams end up with four numbers and no agreement about which is right.

Flux collapses that into a single governed layer that updates itself — and tells you plainly when it couldn’t.

What you get #

Cost that reconciles. Actual and amortized spend, checkpointed daily history, and charge-level detail modelled on FOCUS, the open FinOps billing standard. Sources are cross-checked so the figure on the dashboard is the figure on the invoice.

Inventory that remembers. Every snapshot is retained, so “what changed?” is an exact diff between two points in time rather than a guess from current state.

Utilization you can trust. Metrics reconciled per resource with explicit coverage states. Missing telemetry is never quietly classified as idle — a rule that matters the first time someone proposes deleting a machine.

Findings with evidence. Every recommendation keeps its source, method version, observation window, and confidence, and can be exported as a reproducible evidence pack for a change request.

What makes it different #

Most tools in this space give you a number. Flux gives you the number and the reason to believe it.

The governing idea Every figure traces back to a validated, versioned snapshot. If a source was throttled, stale, or incomplete when that snapshot was taken, the surface showing the number says so — rather than presenting a partial total as an estate-wide one.

That discipline runs through the product. Anomaly detection reports a warming-up baseline instead of firing on thin history. Right-sizing routes conflicting evidence to review instead of picking a winner. Savings are measured from real cost movement after a change, not assumed from the original estimate.

Who uses it #

Team Typical first question
FinOps Which cost centre does this spend belong to, and how much is unallocated?
Cloud operations Why did spend jump on Tuesday, and what changed in the estate?
Engineering Which machines are genuinely oversized, and what evidence says so?
Leadership Where are we against budget, and what have we actually saved?

Where to go next #