.png){kind=avatar}
Accurate interview evaluations depend on four things: a structured, consistent interview process; questions designed around the specific competencies the role requires; real-time support for interviewers during the conversation; and evidence capture that ties what candidates actually said to the competencies being assessed. When any of those are missing, evaluation quality suffers—and hiring decisions end up grounded in impression rather than evidence.
The problem usually isn't that interviewers aren't trying. It's that the conditions required to evaluate candidates accurately are harder to maintain than most teams realize.
Two things break down most often. First, interviewers—especially those who don't interview frequently—struggle to keep conversations on track. Without strong preparation and real-time support, interviews drift. Key competency areas go uncovered. Follow-up questions don't get asked. The conversation becomes more of a general chat than a structured evaluation.
Second, interviewers often don't have great questions going in, or aren't asking consistent questions across candidates. When different candidates get asked about fundamentally different things, there's no reliable basis for comparison. You end up assessing people against different bars without realizing it.
Recent research from Textio makes the consequences of this visible. Across more than 10,000 documented interview assessments, interviewers wrote 39% more feedback for candidates they rejected than for candidates who received offers. That gap holds across interviewer style, role type, and organization. It suggests that people are deciding “no” and then over-explaining to defend it, instead of purely documenting evidence and deciding from there.
The research also found that interviewers write 17% more feedback about women candidates than men—while women are simultaneously more likely to have no documented feedback at all.
And, more than a third of interviewers had commented on a candidate's personality by the time that candidate received an offer. Personality commentary isn't evidence of job performance. It's a signal that the evaluation has drifted away from role criteria and toward impression.
The common thread across all of it: when evaluation criteria aren't defined clearly upfront, instinct fills the gap. The words may multiply, but the signal doesn't improve.
It helps to look at this across the three stages of the interview itself.
Before the interview, the damage is usually done by loosely defined competencies and generic interview questions that aren't anchored to what the role actually requires. When interviewers walk in without clear focus areas and role-specific questions, there's no consistent framework for what they're trying to learn—and no consistent basis for comparing what they find.
During the interview, the problem compounds. Interviewers are splitting their attention between listening, forming follow-ups, and trying to capture notes at the same time. Under that cognitive load, conversations drift. Questions get skipped. Whatever stands out most gets written down — which isn't always what's most relevant to the role.
After the interview, memory takes over. By the time an interviewer sits down to write feedback—especially after a day of back-to-back conversations—the specific things a candidate said have already started to blur. What lingers is the overall feeling. And when evaluation criteria aren't clear, that feeling is what gets documented as an assessment.
Getting consistent, useful feedback out of every interview isn't about asking interviewers to try harder. It's about giving them the right conditions to do it well.
This starts before the first interview. The role needs clearly defined competencies, and each interviewer needs a specific focus area assigned to them—along with role-specific questions tied to those competencies. This is what gives interviewers a concrete framework for the conversation and a clear standard to document against. Generic questions produce generic answers. Role-specific questions produce evidence.
During the interview, interviewers need active support, not just a question list to glance at. That means checklists and real-time prompts that keep the conversation on track and on-competency, and in-the-moment note-taking and evidence capture so what candidates actually say gets recorded as it happens, not reconstructed from memory an hour later.
After the interview, feedback needs to be tied to actual transcript evidence, not impressions. And it needs to be structured in a way that makes it possible to compare what different candidates said against the same role criteria—so the evaluation is grounded in consistent standards across the whole panel, regardless of which interviewer conducted which conversation.
Lavalier is an interview intelligence system built to support evaluation accuracy at every stage. Role Setup aligns hiring managers and recruiters on competencies and evaluation criteria before interviews begin. Plan Builder turns those criteria into structured interview guides with role-specific questions and assigned interviewer focus areas. Live Guidance keeps interviewers on track during the conversation with AI-powered prompts and real-time evidence capture. Candidate Compare maps completed interviews across candidates against the same role criteria, so debriefs are grounded in what candidates actually said rather than what interviewers remember or felt.
Accurate interview evaluations don't come from asking interviewers to be more objective. They come from giving every interviewer clear criteria, the right questions, and real-time support—and from making sure the evidence from every conversation is captured and comparable.
See how Lavalier works across a real role. Try it free today →
