FinOps Maturity Model: Crawl, Walk, Run per Capability
Maturity is assessed one capability at a time, and the FinOps Foundation warns against chasing Run everywhere.
The FinOps maturity model is the FinOps Foundation’s Crawl, Walk, Run scale for describing how well an organization performs a given FinOps capability, such as allocation, forecasting, or anomaly management. The approach is deliberately incremental: start small, take quick action at limited scope, assess the outcome, and grow scale and complexity only as business value warrants.
The most important thing to understand about the model is its unit. It is not a single rating for a company. The Foundation states that every capability and functional activity can be at a different level of maturity, so an organization can be at Run for allocation and at Crawl for commitment management at the same time.
What each level looks like
The Foundation publishes characteristics and sample KPIs from its community for each level. The KPIs below are examples, not thresholds you must hit.
| Crawl | Walk | Run | |
|---|---|---|---|
| Adoption | Capability understood but not followed by all major teams | Understood and followed within the organization | Understood and followed by all teams |
| Process | Basic processes and policies defined | Automation and/or processes cover most requirements | Automation is the preferred approach |
| Edge cases | Plans to address low-hanging fruit | Difficult edge cases identified; some deliberately not addressed (informed ignoring) | Difficult edge cases are being addressed |
| Measurement | Very little reporting and tooling; basic KPIs | Medium to high goals and KPIs | Very high goals and KPIs |
| Sample: cost allocated to a known owner | at least 70% | at least 85% | greater than 90% |
| Sample: commitment discount coverage target | about 60% | greater than 75% | greater than 80% |
| Sample: forecast vs actual variance | below 20% | less than 10% | less than 5% |
Run is not the target
The Foundation is direct about this: an organization’s goal should never be simply to reach Run in every capability. Its example is anomaly detection at Walk that has proven adequate for the few spikes the organization has seen; effort spent pushing it to Run would be better spent on a capability that still has unmet goals. Being at Walk is neither good nor bad in itself. The question is whether the capability is achieving the outcome it exists for.
How to run an assessment
- Pick the capabilities that matter now. Start with the few tied to current pain: unexplained spend, missed forecasts, or unowned cost.
- Define success before rating. Write down what outcome the capability should deliver, then rate against the characteristics above.
- Measure, do not self-report. Allocation coverage, commitment coverage, and forecast variance can all be calculated from billing data. Use numbers where you have them.
- Choose one level-up per quarter. Pick the capability where moving one level returns the most value, and name what changes: a policy, an automation, an owner.
- Re-assess on a fixed cadence, and record the evidence so the next assessment compares like with like.
Evidence a tool can provide
Several sample KPIs come straight out of cost data. In ZopNight, team and tag attribution show how much spend has an owner and how much sits unattributed, and cost anomaly detection runs daily with root-cause analysis on each anomaly. Those give an assessment numbers rather than impressions. What they cannot supply is the organizational side: who follows the process, and which edge cases are consciously ignored.
For the framework itself, see FinOps principles and phases. For capability-level practice, see cost allocation and cost anomaly detection.
Questions we get a lot.
If yours isn't here, email us and we'll answer directly.
Is the FinOps maturity model a company-wide score?
No. The FinOps Foundation applies Crawl, Walk, Run to individual capabilities and domains. Each capability can sit at a different level.
What does informed ignoring mean?
It is a Walk-level characteristic: difficult edge cases are identified and a deliberate decision is made not to address them, rather than leaving them unexamined.
How long does it take to move from Crawl to Run?
The Foundation does not prescribe a timeline, and it advises against aiming for Run in every capability. Progress depends on which capabilities you choose to mature and why.