Move beyond red-yellow-green toward defensible numbers.

In short
Heat maps show where risk sits but not how much it matters. Quantifying cyber-physical risk, even roughly, lets utilities compare mitigations, justify budgets and show regulators why resources went where they did. Start small with high-consequence assets and refine over time.
Most utility risk registers rely on qualitative ratings. They are easy to produce but hard to compare, and they rarely answer the question leadership actually asks: which investment reduces the most risk?
Why quantify cyber-physical risk?
- Compare mitigations on a common scale
- Justify budget requests with numbers, not colors
- Track whether risk is actually going down
- Explain prioritization to regulators and boards
Core inputs to a quantified model
| Input | Example source | Notes |
|---|---|---|
| Asset consequence | Operational impact analysis | Start with high-consequence assets |
| Threat likelihood | Vulnerability and exposure data | Use ranges, not single values |
| Control effectiveness | Assessment results | Credit only verified controls |
| Mitigation cost | Engineering estimates | Enables cost-per-risk comparison |
Four principles for useful risk numbers
Start small
Model a handful of high-consequence assets before scaling.
Use ranges
Honest uncertainty is more defensible than false precision.
Link to evidence
Every input should trace to a record or assessment.
Revisit regularly
Update the model as controls and threats change.
What does this look like in practice?
Cyberwiz-Pro combines asset consequence, vulnerability data and control assessments into risk views that engineers can adjust and explain, all on-premises.
Five questions to start quantifying
- Which ten assets would cause the most impact if compromised?
- What controls protect them today, and are they verified?
- What would each candidate mitigation cost?
- How would you show that risk went down next year?
- Who signs off on the model’s assumptions?
Numbers that support decisions
Risk quantification does not need to be perfect to be useful. It needs to be consistent, traceable and good enough to compare options.
Frequently asked questions
Do we need specialized data scientists?
No. A simple, consistent model maintained by engineers is enough to start.
How accurate are the numbers?
They are estimates with ranges; their value is in comparison and trend, not precision.
Can this support regulatory discussions?
Yes. Traceable inputs make it easier to explain why resources were prioritized.


