What is a bar-raiser interviewer in a hiring loop, and what weight does their scorecard carry?

## The role Most panel slots are staffed by the team doing the hiring: future teammates, a neighbouring team, the manager. A **calibrating interviewer** — the role widely known in the industry as a *bar raiser*, and copied under many local names — is deliberately not one of them. They are pulled from a different part of the organisation, they have interviewed across many loops, they are trained for the role, and they have no stake in whether this particular seat is filled this quarter. The practice is documented and unremarkable at scale. What varies between employers is the name, the training pipeline, and exactly how much formal power the role holds. What is consistent is the shape: one round of the loop is run by an independent interviewer whose write-up is weighted differently from everyone else's. ## Why the role exists at all Hiring standards drift, and they drift in one direction. A team that is understaffed against a deadline is asking *is this person good enough to help now*. That is a rational question and a corrosive one, because the organisation needs the answer to a different question: *is this person good enough that the average goes up, and would this hire still look strong in three years, possibly on another team entirely*. Nobody decides to lower the bar. It slips one reasonable exception at a time. The calibrating interviewer is the structural response: an outsider whose only job is the durable standard, plus a written scorecard whose weight is agreed **before** the loop rather than argued after it. ## The scorecard and where the weight lives Every interviewer files a scorecard: what was asked, what the candidate said or did, a rating, a recommendation. The calibrating interviewer files one too, but it is read differently in the debrief. It is expected to answer whether the candidate clears the durable standard, at which rung, and on what evidence — and in many processes it can halt a hire the team is enthusiastic about. That is what veto weight actually means in practice. Not that the calibrator argues more loudly, and not that they outrank anyone. It means an offer does not proceed over a blocking write-up unless new evidence appears or someone escalates explicitly. Five scorecards leaning hire do not out-vote it, because it was never a vote. A blocking recommendation is also not the only way the write-up moves an outcome. A hedged one — *clears the bar, but the evidence for the posted level is thin* — routinely produces a lower rung or an extra round instead of an offer. The recorded doubt does the work. ## What the round looks like from the candidate's chair Here is an invented but typical shape. A platform-infrastructure candidate sits a six-round onsite. The fifth slot is 45 minutes with an engineer from an organisation that has nothing to do with the role. Roughly eleven minutes trace the arc of the candidate's last few years; roughly twenty-six go deep inside a single piece of work — decisions, alternatives rejected, what failed, what they would change; about eight minutes remain at the end. From the inside it feels like an ordinary round, sometimes a friendlier one. The difference is entirely in the write-up. The calibrator shares no context with the panel, so everything the team already knows has to be re-established from scratch, and anything the candidate leaves implicit does not arrive as *assumed* — it arrives as *absent*. ## The failure this round punishes The single most common mistake is handing the calibrating interviewer the same answers the peer rounds already got. It hurts twice. First, the story was tuned for someone who shares your vocabulary and your context. An outsider needs the scope, the constraints, the stakes and the counterfactual stated plainly, and a story compressed for insiders reads as thin to them. Second, the debrief reads the scorecards side by side. A candidate who brings one project and tells it three times produces a packet with a single data point in it, and the discussion that follows is about breadth rather than about the offer. ## Preparing for it - Assume one round is calibrating and that you will not reliably know which. - Make each round's evidence self-contained: scope, constraints, timespan, what you owned versus influenced. - Budget distinct material across the day rather than polishing one story to a shine. - Answer the *why* behind decisions, not only the *what* — standard-relative probing sounds like *compared to what* and *what would have happened otherwise*. - Expect the friendly outsider to be the toughest write-up of your day, and be glad the standard exists when you are the one already inside it.