How to reduce noise

What is noise and how does it impact insurances?

Noise is usually defined as an unwanted sound that is considered unpleasant, loud, or disruptive to hearing. From a physics standpoint, there is no distinction between noise and desired sound, as both are vibrations through a medium, such as air or water. However, Daniel Kahneman defines noise as an “undesirable variability in judgments of the same problem in the context of human judgment and decision-making.

In his most recent nonfiction book “A Flaw in Human Judgment” written together with Olivier Sibony and Cass Sunstein, the authors focus on the issue’s statistical properties and psychological perspectives.

Examples they give include two psychiatrists who independently diagnosed 426 state hospital patients and only in half of the cases agreed on which mental illness the patient suffered from.  Another finding showed that French court judges were more lenient if it happened to be the defendant’s birthday.

And finally, a finding at an insurance company showed that the median premiums set by underwriters independently for the same five fictive customers varied by 55 %, five times as much as expected by most underwriters and their executives.

All cases are exciting observations of a more rampant problem than we might think. It’s tolerated not because it’s acceptable but because it remains unnoticed. Without monitoring noise, many organizations can be fooled into an illusion of agreement while disagreeing with their judgments. We can live comfortably with colleagues without ever noticing that they do not see the world as we do.

“Wherever there is judgment, there is noise—and more of it than you think.” – Daniel Kahneman, Olivier Sibony, Cass R. Sunstein (2021) Noise: A Flaw in Human Judgment

Why is noise a problem in organizations?

Noise is an ignored problem in organizations that leads to bad judgments. Bias gets all the limelight but noise also creates problems. To understand the error in judgment, we must understand both bias and noise. In professional judgments of all kinds, whenever accuracy is the goal, bias and noise play the same role in the calculation of overall error.

Noise can be wanted and unwanted, and the amount of unwanted noise is greater than most organizations realize. It creates inconsistency, and inconsistency damages the credibility of the system and connected services or products. Consequently, it can have a great cost both financially and socially.

For example, in criminal punishment, we expect similar sentences for similar crimes. If a group of judges gives different sentences to defendants who committed the same crime depending on the judge then one could call the system noisy.

How can we reduce noise in claims valuation?

Thinking of the insurance industry, it becomes obvious that there is a lot of noise clouding judgment, impacting decisions that might result in perceived inconsistency from the policyholders’ perspective. Using accurate, data-driven systems for subjective tasks like valuation has the potential to vastly reduce noisy judgments. Especially when these valuations are large in numbers and involve technically complex objects such as mobile phones or computers.

That’s why insurance-related processes such as claims valuation present a clear application area for automation. Improving the quality of valuation and decision-making should be a no-brainer to reduce any inconsistencies in the perspectives of the policyholder.