The Data They Didn’t Have Held The Key

In a recent conversation with Decoded Thinking, contact centre consultant Paul Weald reflected on a World War Two story that still feels surprisingly relevant to modern AI. During the war, military analysts studied returning aircraft to understand where enemy fire caused the most damage. The assumption seemed logical: reinforce the areas with the most bullet holes.

But statistician Abraham Wald saw the problem differently.

The military was only analysing the planes that made it home. The aircraft that had been hit in more critical areas never returned at all.

Today, the lesson is known as survivorship bias and it remains one of the clearest explanations of how incomplete datasets can distort decision-making.

As AI becomes more embedded into customer insight, healthcare, recruitment, policing, and operational systems, the question is no longer simply what data organisations can analyse.

It is what they might still be missing.

Read the full Thinking piece