The longer you look, the less you see

By Kirill Ivanov, co-founder and CEO

The longer you look for a mistake, the less likely you are to find it.

Psychologists call it the vigilance decrement. It has been replicated for over 75 years.

During World War II, concerns about radar operators missing submarine signals prompted psychologist Norman Mackworth to study sustained attention. In experiments published in 1948, participants missed more signals after the first half-hour of a monotonous monitoring task.

Photo: WAAF radar operator Denise Miley at Bawdsey Chain Home station, May 1945. Flt Lt Goodchild, Royal Air Force official photographer. Imperial War Museums (CH 15332), public domain, via Wikimedia Commons.

Label review presents a similar challenge. Hundreds of details are correct, one might be wrong, and someone has to find it on every label, every day.

Our clients tell us this all the time. Catchmaster, for example, shared that five people can look at the same label and only one of them will spot the error. Even with several people reviewing, mistakes can still slip through.

That’s why I believe label review is one of the best use cases for AI. I don’t want AI designing packaging, coming up with product ideas or deciding how a brand should speak to its customers. Those need taste and a personal touch. I want it to take over work that is boring but critical.

According to Sedgwick’s analysis, undeclared allergens were the leading cause of FDA food recall events in the first half of 2026. A single food recall can cost millions, before accounting for lost sales and damage to the brand.

Let AI carry that load, so people can focus on the work that needs judgment and creativity.

In my next post, I’ll share why consistency is what brands need most in label review, and why general AI tools still struggle with it.