Risk mitigation
Risk mitigation means taking action to reduce the likelihood of a harmful event, its impact, or both. In business, it means putting safeguards in place against risks such as lost revenue, operational failures or data loss.
How it works
Identify a specific risk, estimate its likelihood and impact, then choose a safeguard. Assign someone to maintain it and check whether the remaining risk is acceptable.
Start with the largest threats, not the easiest fixes. For a startup, a risk that could exhaust its cash deserves attention even if it is unlikely.
Backups cut expected annual costs by $3,500
A hypothetical software startup estimates a 10% annual chance of a data-loss incident. Without backups, recovery would cost $50,000. A backup plan costs $1,000 annually and reduces the estimated loss per incident to $5,000.
For simplicity, the model allows at most one incident per year.
| Annual calculation | No backup plan | With backup plan |
|---|---|---|
| Expected incident loss | 10% × $50,000 = $5,000 | 10% × $5,000 = $500 |
| Backup plan cost | $0 | $1,000 |
| Total expected cost | $5,000 | $1,500 |
The expected annual saving is $5,000 − $1,500 = $3,500. The plan reduces the impact, not the probability, of an incident.
Expected cost is a planning measure, not a forecast. The startup still needs enough cash and a recovery process to survive an actual incident.
Questions people ask
What are the four types of risk mitigation?
People often mean four common risk responses: avoidance, reduction, transfer and acceptance. Avoidance means stopping the risky activity; reduction means lowering likelihood or impact; transfer shifts some exposure through arrangements such as insurance; acceptance means choosing to bear the risk. Strictly, mitigation means reduction, not all four.
What are risk mitigants?
Risk mitigants are the safeguards themselves: backups, spending limits, supplier checks or access controls. A useful mitigant addresses a named risk and has an owner. "We will be careful" is not a control you can test.
