You Built a Structured Interview Scorecard. Your Hiring Managers Aren't Using It.

The scorecard is filled out. Every box has a number. The debrief runs for twenty minutes and ends with consensus. You hire the candidate. Six months later you're having a performance conversation you could have scripted in advance, and somewhere in the back of your mind you're asking why the process didn't catch this. It did catch it. You just didn't listen to it.

This is not a story about bad scorecards. The scorecard was probably fine. The problem is everything that happens around it, specifically the social dynamics of an interview loop that quietly hollow out structured process until what remains is a paper trail for a decision that was already made on instinct.

What "structured" actually means, and why it erodes so fast

A structured interview has three properties: the same questions asked in the same order to every candidate, answers scored against a defined rubric before discussion, and scores recorded independently before any debrief. Remove any one of those three and you no longer have a structured interview. You have a slightly more organised unstructured interview, which is not the same thing.

Most sales teams remove all three without realising it. Questions drift because the conversation "goes somewhere interesting." Scores get filled in retrospectively, during or after the debrief. And independent scoring is treated as a formality rather than the actual protection it provides.

The protection it provides is this: it forces each interviewer to commit to an assessment before they know what everyone else thinks. Once you know what the VP of Sales thought of the candidate, your own recollection of the interview starts to bend towards theirs. That is not weakness; that is neuroscience. The process is supposed to protect you from it, and most processes don't.

Where it actually breaks down: three named failure modes

The logo override. A candidate comes in with a recognisable name on their CV, Salesforce, Gartner, a well-regarded SaaS scale-up. Before they've answered a single question, the room has a prior. The structured process asks you to score what you observed. The logo tells you what you expect to observe. Confirmation bias does the rest. On the scorecard, this shows up as inflated scores on competencies that were never properly tested, because the interviewer felt the answer was strong without being able to say exactly why.

The charisma collapse. This one is more insidious because it mimics competence. A candidate who is fluent, confident, and socially skilled will fill silences with credible-sounding content. Interviewers who haven't been trained to probe for specifics will accept fluency as evidence. The rubric might say "demonstrates structured approach to discovery," but in practice the interviewer is scoring their feeling of being impressed. Those are not the same thing. I have sat in debriefs where a candidate scored fours across the board and nobody could tell me what the candidate actually said when asked to describe a specific deal.

Seniority collapse in the debrief. This is the one nobody wants to name. You have four interviewers. Three of them fill in their scorecards. The fourth is the hiring manager, or the VP, or the founder. They speak first in the debrief. Within four minutes, the other three scores have migrated towards theirs. Not because anyone changed their mind on the evidence. Because the social cost of disagreeing with a senior person about a decision they clearly want to make is higher than the abstract benefit of maintaining an independent position on a hiring rubric. The independent scores that existed five minutes ago no longer exist in practice.

The fix is procedural, not cultural

You will not solve this by asking people to "be more rigorous" or "trust the process." That is change management theatre. The interventions that actually work are mechanical, and they work because they remove the moment of choice.

Here is a short set that holds structure intact without requiring anyone to become a psychometrician or to feel like their judgement is being questioned.

1. Scores submitted before the debrief, to a neutral collector. Not filled in at the debrief. Not filled in on the way to the debrief. Submitted, in writing, to a recruiter or coordinator, before anyone speaks. This is the single highest-leverage change you can make. It costs nothing except a small amount of coordination. If a hiring manager resists this, that resistance is itself useful information.

2. The debrief starts with divergence, not consensus. The coordinator shares a summary of where scores differ before discussion begins. The conversation starts with: "There's a two-point gap on commercial acumen between X and Y. What did you each observe?" This reframes the debrief as a calibration exercise rather than a ratification exercise.

3. Anchored rubrics, not scale descriptions. A rubric that says "4 = strong performance" is worthless. A rubric that says "4 = candidate named a specific stakeholder they had to re-engage mid-cycle, described the tactic they used, and could articulate what changed as a result" is testable. Build your rubrics around observable, specific behaviours. If you don't have these, your scorecard is just a voting slip. The Sales Skills Self-Assessment framework we use at Peer is built on exactly this principle: every competency is anchored to a behaviour, not a trait.

4. The hiring manager scores last, not first. In the debrief, the most junior interviewer speaks first. This is uncomfortable the first two or three times. After that it becomes normal, and the quality of independent input improves significantly. You will hear things the hiring manager missed. That is the point.

5. A mandatory "what specifically did they say" question for any score above a three. Before a high score stands, the interviewer who gave it has to quote something specific from the candidate's answer. Approximate paraphrase is acceptable. "They were just really impressive" is not. This rule alone eliminates a significant proportion of logo and charisma inflation, because it surfaces the absence of real evidence in a way that is hard to dismiss.

What you're actually asking people to do

These interventions feel like bureaucracy if you describe them badly. Frame them correctly and they are something simpler: a way to make sure that the hiring manager's instincts get tested rather than just ratified.

The best hiring managers I've worked with welcome this. They know their instincts are good and they also know their instincts are occasionally catastrophically wrong, and they would rather have a process that catches the latter than a process that flatters the former. If a hiring manager tells you the structured process is slowing them down, ask them what their 90-day attrition rate looks like. That usually ends the conversation.

The scorecard was never the problem. It was always the thirty seconds after the candidate left the room, when someone senior said "I loved them," and everyone else quietly forgot what they wrote down.