The four-stage loop that stops good candidates walking

Recruiter screen for motivation and logistics, hiring manager for scope and judgement, a technical or craft round for how someone actually thinks, and a team or founder conversation for fit in both directions. Past that, additional rounds are usually testing the same thing twice.

Each of those four stages is answering a genuinely different question, which is why none of them is safe to cut. The recruiter screen filters on logistics and motivation before anyone's calendar is spent — location, notice period, comp expectations, why they're looking. The hiring manager round is where scope gets tested against experience: has this person actually done something close to what the role needs, at a comparable level of ownership. The technical or craft round is the only stage that should be evaluating how someone thinks under real conditions, not how well they've rehearsed an answer. And the team or founder conversation is doing something the earlier three can't: letting the candidate evaluate you back, which matters more the more senior the hire is.

The instinct behind a fifth round is usually good: someone on the panel is unsure, and another data point feels like the safe move. In practice it rarely resolves the uncertainty — it just moves it to a different interviewer, who is now evaluating the same competency the earlier panel already covered, with less context and more fatigue than the people who saw the candidate first.

What a fifth round reliably does is add ten to fourteen days to a process that was already at risk of losing the candidate to a competing offer. We've seen strong candidates accept a faster competing offer specifically because the additional round read as indecision, not thoroughness — and from the outside, those two things look identical.

We track loop length against three outcomes across every search: withdrawal rate, time-to-offer, and — where clients share it — 12-month retention of the hire. The pattern is consistent enough that we now flag any loop proposed at five stages or more before a search opens.

If two people on your team are evaluating the same competency — say, two engineers both assessing technical depth — put them in the same session instead of scheduling separately. It halves the calendar cost without losing any signal, and it often produces a better read, because the two interviewers can compare notes on the same answer in real time instead of reconciling separate impressions after the fact.

The same logic applies to culture and values assessments that get bolted onto an existing round rather than given their own slot. A well-run team conversation already tells you most of what a separate “values fit” interview is trying to establish — adding a dedicated round for it is usually redundant, not additive.

The four-stage shape holds up across functions, but what each stage is actually testing shifts. For an engineering hire, the technical round is usually the highest-signal stage and the one worth protecting most carefully from being rushed or combined with something else. For a go-to-market hire, the equivalent high-signal stage is often a live exercise — a mock pitch, a discovery call role-play — rather than a Q&A round, because the skill being assessed is inherently behavioral, not something a candidate can describe accurately about themselves.

Executive searches are the one place we'll defend a longer loop as standard rather than exceptional, because the cost of a bad senior leadership hire is high enough that additional stages — meeting the board, presenting to the wider team — are answering genuinely different questions, not re-testing the same one. Even there, though, the same rule applies: every stage should be nameable, or it shouldn't be on the calendar.

The request for a fifth round almost always comes from a good place — a genuine desire to get the hire right, especially for a role that feels consequential. The useful response isn't to refuse outright, it's to ask what specific question the new round would answer that the existing four didn't. Most of the time, the honest answer is “I'm not fully sure, I just want more information,” which is a reasonable feeling but not a basis for an extra round — it's a basis for reviewing the existing scorecards more carefully before deciding.

Where we do support a fifth stage is when it's answering something genuinely new — a domain-specific technical deep dive for a highly specialized role, for instance, where the standard technical round couldn't reasonably cover the full scope. The distinction is whether the round tests something new, not whether it makes the panel feel more confident. Confidence built on redundant information is false confidence, and it comes at a real cost in withdrawal risk.

The four-stage structure works best when it's agreed before a search opens, not improvised stage by stage as a candidate moves through. Loops that get extended mid-process — because a stakeholder wants in, or because one interviewer was unsure — are the ones most likely to blow past four stages without anyone deciding that on purpose.