Tools › Project Management › Risk › Cyber Risk Quantification › Example
Three cyber loss scenarios and a control: expected loss per year, net benefit and the loss exceedance curve before and after
| Tool | Cyber Risk Quantification |
| Source | original |
| Level | Professional |
Problem
A small firm lists three loss scenarios as 90 % ranges: ransomware (0.05 to 0.4 events a year, 40,000 to 600,000 per event), e-mail payment fraud (0.2 to 1.5 a year, 5,000 to 120,000) and an outage of a key system (0.5 to 3 a year, 2,000 to 40,000). A control costing 12,000 a year would prevent 60 % of the ransomware and payment-fraud events. Find the expected loss per year before and after, the net benefit and the return on the control's cost, and the loss with a 10 % and a 1 % chance of being exceeded in a year.
Inputs to Enter
- Scenarios (events per year low / high, loss per event low / high, covered by the control): Ransomware 0.05 / 0.4, 40,000 / 600,000, yes; E-mail payment fraud 0.2 / 1.5, 5,000 / 120,000, yes; Outage of a key system 0.5 / 3, 2,000 / 40,000, no
- Control: cost 12,000 a year, 60 % fewer events, losses not smaller
- Simulation: 10,000 years, seed 1, uncertain event rate (the defaults)
Answers
- Ransomware loss per event: μ = (ln 40,000 + ln 600,000) / 2 = 11.951, σ = (ln 600,000 − ln 40,000) / (2 × 1.645) = 0.8232, mean = exp(μ + σ²/2) = 217,396.85 (median 154,919)
- Ransomware events per year: σ = 0.6321, mean = √(0.05 × 0.4) × exp(σ²/2) = 0.1727; expected loss 0.1727 × 217,396.85 = 37,543.17 a year
- Expected loss per year, closed form: 37,543.17 + 25,808.00 + 19,234.43 = 82,585.60 (average of the simulated years 82,718, standard error 1,366)
- After the control: 0.4 × (37,543.17 + 25,808.00) + 19,234.43 = 44,574.90; reduction 38,010.70; net benefit 38,010.70 − 12,000 = 26,010.70 a year; return 217 % of the cost
- Simulated (seed 1): 10 % chance of a year above 220,232 before and 108,313 after; 1 % chance of a year above 666,425 before and 437,765 after; a loss in 85 % of years
- Ransomware is 45 % of the expected loss and 92 % of the loss in the worst 1 % of the simulated years
Notes
The tool opens on this case; press Submit, or open it in the tool: /analysis-tools/project-management/risk/cyber-risk-quantification/?try=cheap-control-big-cut. Every number is an original illustrative figure, not an industry benchmark. Closed forms checked against scipy and the simulated figures against an independent numpy simulation of 2,000,000 years (10 % loss 209,502, 1 % loss 682,101 before the control), within sampling error. The output is only as good as the entered ranges.