Tools › Project Management › Risk › Cyber Risk Quantification › Example

A control that costs more than the expected loss it removes, and what it does to the 1 % year

ToolCyber Risk Quantification
Sourceoriginal
LevelProfessional

Problem

Two scenarios: a customer data breach (0.02 to 0.2 events a year, 100,000 to 1,500,000 per event) and a lost or stolen laptop (0.5 to 4 a year, 1,000 to 15,000). A service costing 60,000 a year would make each breach 40 % smaller and does not act on the laptops. Compare its cost with the expected loss it removes and with the change in the loss that has a 1 % chance of being exceeded.

Inputs to Enter

  • Scenarios: Customer data breach 0.02 / 0.2 events a year, 100,000 / 1,500,000 per event, covered; Lost or stolen laptop 0.5 / 4 a year, 1,000 / 15,000, not covered
  • Control: cost 60,000 a year, events not fewer, losses 40 % smaller
  • Simulation: 10,000 years, seed 1, uncertain event rate
Open the Cyber Risk Quantification tool

Answers

  • Breach: mean 0.0808 events a year × mean loss 543,492.11 = 43,914.30 a year; laptops: 1.7269 × 5,434.92 = 9,385.79 a year
  • Expected loss per year 53,300.10 before and 0.6 × 43,914.30 + 9,385.79 = 35,734.37 after; reduction 17,565.72
  • Net benefit 17,565.72 − 60,000 = −42,434.28 a year; return −70.7 % of the cost: each 1.00 spent removes 0.29 of expected loss
  • Simulated (seed 1): the loss with a 1 % chance of being exceeded falls from 1,051,366 to 634,557; the 10 % loss barely moves (38,477 to 38,465) because nine years in ten have no breach
  • The breach is 82 % of the expected loss and nearly all of the loss in the worst 1 % of years

Notes

Open this case in the tool: /analysis-tools/project-management/risk/cyber-risk-quantification/?try=control-costs-more-than-it-saves. Original illustrative numbers. The tool reports both the average and the tail figures; whether a thinner tail is worth the net cost depends on how much loss the organization can absorb, which the tool does not judge. Independent numpy simulation: 1 % loss 1,055,281 before and 637,669 after.

More Examples for This Tool