Why Interviews Mislead

The Core Idea

Unstructured interviews mostly measure rapport between two people in a room, not the candidate's ability to do the job. Structured interviews replace conversation with a pre-written scoring rubric, so the decision rests on comparable evidence rather than mood.

Why Comfort Crowds Out Skill

When no rubric exists, the interviewer fills the silence with whatever signals feel reassuring in the moment. A candidate who mirrors the interviewer's humor, shares an alma mater, or simply seems relaxed scores well for reasons that have nothing to do with the role. Comfort and competence overlap, but only partly, and the gap is where bad hires come from. Once a decision feels right in the body, the brain tends to retrofit reasons for it, which is why gut hires are remembered as wins even when the track record is mixed.

What a Structured Interview Actually Looks Like

The format is mechanical on purpose. Before any candidate arrives, the hiring team writes a list of the job's core skills, then writes one question per skill. Each question comes with a scoring scale, often 1 to 5, anchored by short descriptions of what a 1, 3, and 5 answer looks like. Every candidate gets the same questions in the same order, scored against the same anchors. Two interviewers scoring the same transcript independently usually agree closely, which is the property unstructured chats lack.

A worked example for a junior analyst role:

• Skill: Spotting a flawed assumption in a dataset.
• Question: Given this sales table, identify the row where the reported growth is most misleading and explain why.
• Anchor for a 3: Correctly names the row and cites the denominator change.
• Anchor for a 5: Names the row, explains the denominator, and suggests the corrected metric.

Friendliness adds no points. A charming candidate who spots the wrong row scores a 1.

The Common Misunderstanding

Many people hear "structured" and picture a script read aloud like a robot, then assume the format kills the signal a thoughtful interviewer would otherwise catch. The opposite is usually true. The rubric forces the interviewer to write down, before meeting anyone, what a good answer actually looks like. That work, done once, replaces dozens of vague impressions collected over months.

When Structure Does Less

Structured interviews are well suited to roles where the work can be sampled in a short exchange, such as customer support, junior analysis, or technical writing. They are weaker for senior roles where the sample question cannot approximate the actual decisions the hire will make, and for creative or relational work where the artifact is the person rather than a deliverable. In those cases, work samples, trial projects, or panel reviews across multiple candidates tend to carry more signal than any interview format.

The Monday Morning Step

Write the scorecard before posting the role. List three to five skills the job genuinely requires, attach one question and one rubric to each, and require every interviewer to score independently before discussing the candidate. Comparing numbers is faster and fairer than comparing memories.