The FinOps maturity model in practice.

This article turns the model into something you can act on this quarter: what each stage actually looks like, how to place yourself, and how to choose the next investment.

What do crawl, walk, and run actually look like?

The stage names are easy to recite and hard to apply, because they describe behaviour, not tooling. Crawl is reactive: someone looks at the bill when it spikes, savings happen in bursts after a shock, and cost is finance's problem rather than engineering's. There is visibility somewhere, often a dashboard few people open, but no routine and no ownership. Many organisations with significant cloud spend are still here on most capabilities, and the bill grows faster than the business because nothing holds it.

Walk is consistent. Cost is allocated to teams and products reliably, a regular cadence reviews spend against forecast, and named owners are accountable for their lines. Commitments are sized to a forecast rather than guessed, and engineering sees its own cost. This is where most of the durable savings are captured, because the discipline is repeatable rather than heroic. The shift from crawl to walk is mostly organisational: ownership, cadence, and trustworthy allocation, not a new platform.

Run is continuous and largely automated. Rightsizing happens on a loop, commitment coverage is managed actively, cost shows up in engineering workflows and even in deployment decisions, and anomalies are caught early by detection rather than at month end. Run is expensive to reach and to sustain, which is exactly why it should be reserved for the capabilities and the spend where the return justifies it.

Why measure maturity per capability?

A single maturity grade hides the truth and misdirects investment. An organisation can run a sophisticated commitment program, genuinely at run on commitment management, while still being at crawl on cost allocation, so no team knows what it spends. Grading the whole organisation at, say, walk averages those out and hides both the strength and the gap. Assessing each capability separately exposes the real shape: where you are strong, where you are weak, and therefore where the next dollar of effort returns the most.

The capabilities worth scoring separately are visibility, allocation, forecasting, commitment management, and engineering accountability. Visibility is whether spend is seen, in detail, by the people who can act. Allocation is whether every cost has an owner. Forecasting is whether you can predict spend well enough to commit and to budget. Commitment management is whether coverage tracks a risk adjusted forecast across AWS Savings Plans and Reservations, Azure Reservations and the Azure Savings Plan, GCP Committed Use Discounts, and OCI Universal Credits. Engineering accountability is whether the teams creating cost are answerable for it.

Table: what each stage looks like across the five capabilities worth scoring separately.

How do you place your organisation honestly?

Score each capability against the descriptions above using evidence, not aspiration. The test for visibility is whether an engineering lead can see their own spend without asking finance. The test for allocation is what share of the bill is reliably attributed to an owner, with untagged spend as the inverse measure. The test for forecasting is how far recent forecasts have landed from actuals. The test for commitment management is whether coverage is tied to a defensible forecast or to last year's run rate. The test for engineering accountability is whether a team that doubles its spend hears about it from its own dashboard or from a surprised CFO.

The output is a profile, typically uneven, and that unevenness is the useful part. It usually shows one or two strong capabilities masking weak foundations, and the weak foundations are where the next savings live. Resist the urge to round the profile into a single comforting number, because the number is precisely the information you need to throw away.

Which move pays next?

Sequence matters because the capabilities depend on each other. Allocation underpins accountability, because no team can own a cost it cannot see, and forecasting underpins commitment management, because coverage is only as good as the forecast behind it. So a strong commitment program built on weak forecasting is fragile, and an accountability push built on weak allocation will stall. The highest return move is usually the foundational capability that is holding the others back, even when a flashier one looks more appealing.

Weigh each candidate move by the spend it touches and the savings it unlocks, not by how advanced it sounds. Pushing allocation from crawl to walk on a large, opaque estate typically returns more than pushing an already strong commitment program from walk to run. And accept that some capabilities should stay at walk, or even crawl, where the spend does not justify the cost of automation. The model is a guide to where effort pays, not a checklist to complete.

How fast should maturity move?

Maturity is earned at the speed an organisation can absorb change, not bought in a quarter. The crawl to walk shift is mostly about people and process, and it can move quickly because it does not wait on heavy tooling. The walk to run shift leans on automation and on engineering culture, and it moves more slowly because it changes how teams work. A realistic plan advances one or two capabilities at a time, banks the savings, and uses them to fund the next step, rather than attempting a wholesale transformation that stalls under its own weight.

Where this fits

The maturity model is the map; the rest of the FinOps foundations cluster is the terrain. For what the discipline is and is not, see what FinOps actually is and what it is not, for the structure that supports it see the FinOps operating model explained, and for the phased loop behind run see inform, optimize, operate, the FinOps phases. The estate wide playbook these capabilities serve lives in the cross cloud cost optimization guide.

Find your stage and your next move

We assess your FinOps maturity capability by capability, then sequence the moves that lower the bill in the order that pays, with zero provider commissions. Our guarantee: we reduce your cloud spend or we reimburse our service fee. Pricing is either a Fixed Fee scoped up front or Gainshare, a share of verified savings with no retainer and no risk.