3.3 Leading with Critical Thinking

Key Takeaways

  • Critical thinking is disciplined, evidence-based reasoning: define the problem, gather and test information, weigh alternatives, and check your own assumptions.

  • Common biases in insurance decisions include confirmation bias, anchoring, availability bias, overconfidence, sunk-cost thinking, and groupthink.

  • Good evidence is relevant, reliable, sufficient, and current; a single loss or anecdote rarely justifies a portfolio-wide change.

  • Separating facts, assumptions, and opinions, and asking what would change the conclusion, keeps analysis honest.

  • Leaders show critical thinking by making their reasoning visible so others can test it, not by being the first to answer.

Last updated: September 2026

Leading with Critical Thinking

Quick Answer: Critical thinking is the disciplined habit of reasoning from evidence rather than impulse. In practice, you define the real problem, gather and test information, identify assumptions and biases, compare alternatives against clear criteria, and reach a conclusion you can explain and defend. In insurance, it separates a sound underwriting, claims, or risk management recommendation from one driven by the last big loss or the loudest voice.

Why It Matters in Risk Management and Insurance

Insurance professionals decide under uncertainty every day. They choose whether to accept a submission, how to reserve a claim, which risk control a client should fund, and whether a coverage question needs a reservation of rights. These decisions carry large dollar consequences. They are also easy to distort with shortcuts. CPCU treats critical thinking as a leadership skill because it is what lets a professional recommend, defend, and adjust a course of action.

A Practical Critical-Thinking Process

  1. Define the problem precisely. Ask what decision is actually needed. "Our auto book is losing money" is a symptom. "Losses on fleets with more than 20 vehicles rose sharply in two territories" is a problem you can analyze.
  2. Gather relevant information. Pull loss runs, inspection reports, exposure data, contract terms, and input from the people closest to the issue.
  3. Evaluate the evidence. Test whether each piece of information is reliable, current, and sufficient.
  4. Identify assumptions and biases. List what you are taking for granted. Ask who benefits from each conclusion.
  5. Generate and compare alternatives. Look for at least three options, then weigh cost, effectiveness, feasibility, and side effects.
  6. Decide and explain. State the recommendation, the reasons for it, and the evidence that would change it.
  7. Reflect after results arrive. Compare outcomes with expectations and update your approach.

Separating Facts, Assumptions, and Opinions

TypeDefinitionInsurance example
FactVerifiable informationThe insured has had three water-damage claims in four years
AssumptionSomething accepted as true without proof"The new plumbing contractor fixed the root cause"
OpinionA judgment or belief"This is a well-run building"
InferenceA conclusion drawn from factsRecurring claims suggest an unresolved plumbing or maintenance issue

A useful discipline is to label each statement in a referral or recommendation. Decisions resting mainly on assumptions and opinions need more verification before they go forward.

Biases That Distort Insurance Decisions

  • Confirmation bias: Looking only for evidence that supports your first view. An underwriter who likes an account may skim past a negative inspection note.
  • Anchoring: Relying too heavily on the first number seen, such as last year's premium or the claimant's opening demand.
  • Availability bias: Overweighting recent or vivid events. After one large fire, every similar account can look dangerous.
  • Overconfidence: Believing your estimate is more precise than it is, such as setting a case reserve without allowing for litigation risk.
  • Sunk-cost thinking: Continuing a failing program because of what has already been spent on it.
  • Groupthink: Going along with the team to avoid conflict, so no one challenges a weak plan.

Countermeasures: Seek disconfirming evidence on purpose. Use a checklist. Assign someone to argue the other side. Compare your estimate with base rates. Write down your reasoning before seeing others' opinions.

Evaluating Evidence

Ask four questions of any data point:

  1. Relevance: Does it bear on this decision? A company-wide loss ratio says little about one class of business.
  2. Reliability: Where did it come from? Was it verified? An insured's self-reported sprinkler system differs from an engineer's inspection.
  3. Sufficiency: Is there enough of it? One claim is an anecdote; a credible loss history is evidence.
  4. Currency: Is it up to date? A five-year-old inspection may no longer describe current operations.

Worked Scenario: The Warehouse Renewal

An underwriter must decide whether to renew a distribution warehouse after a $400,000 fire. The branch manager wants to non-renew immediately.

  • Define: The question is whether the account is still acceptable at an adequate price, not whether the fire was bad.
  • Gather and evaluate: Loss control confirms the fire started from an improperly stored battery charging station. The insured has since moved charging to a separate, sprinklered room. The prior 10 years show no other property losses.
  • Check biases: The urge to non-renew reflects availability bias from the recent fire. The broker's insistence that "it will never happen again" reflects overconfidence.
  • Alternatives: Non-renew; renew as is; or renew with a protective safeguards condition, a higher deductible, and an adjusted rate.
  • Decide and explain: Renew with conditions. Document the root-cause evidence and name the facts that would change the decision, such as noncompliance at the next inspection.

Common Traps

  • Treating speed as decisiveness. Fast decisions are good only when the evidence is clear.
  • Confusing correlation with causation, for example, assuming a rating variable causes losses rather than simply predicting them.
  • Stopping at the first acceptable alternative instead of comparing several.
Loading diagram...
Critical Thinking Loop
Test Your Knowledge

After one large liability verdict against a single restaurant, a manager proposes declining all restaurant submissions. Which bias is most likely at work?

A

Anchoring

B

Sunk-cost thinking

C

Confirmation bias

D

Availability bias

Test Your Knowledge

An underwriter's referral states, 'The insured's new roof eliminates the leak problem.' No inspection or invoice supports the statement. How should this statement be classified?

A

An assumption that needs verification

B

A verified fact

C

An inference drawn from inspection data

D

An irrelevant opinion

Test Your Knowledge

Which step best protects a team against groupthink when evaluating a new program?

A

Letting the most senior person state a view first

B

Limiting the discussion to members who supported similar programs before

C

Deciding quickly so that disagreements do not slow the launch

D

Assigning someone to argue against the proposal and inviting written views before discussion

Sections you finish are checked off in the contents.