The Gut-Feel Penalty
Structured interviews predict job performance about twice as accurately as the free-flowing kind. The difference isn't the questions. It's whether every candidate answers the same ones.
A relaxed, free-flowing interview feels like the one where you really get to know a candidate. It also predicts how they'll perform about half as well as a structured one.
The most recent large meta-analysis of hiring methods puts the validity of structured interviews at 0.42, versus 0.19 for unstructured (Sackett et al., 2022). Roughly twice the predictive power, from the same hour in the same room.
The mechanism is comparison. When every candidate answers different questions, you have no common yardstick, so you fall back on rapport, similarity, and confidence (none of which predict performance). When everyone answers the same questions against the same scoring guide, you're measuring the candidate instead of the conversation.
Structured doesn't mean robotic. It means fixing the questions before you meet anyone, defining what a strong answer looks like, and scoring each response as you go. The warmth can stay. The yardstick is what's new.
The takeaway: the interview that feels the most natural is usually the least predictive.
