Interview design
Draw the line, then check it.
A pass mark decides who you read first. Pick a sensible starting number, calibrate it on your first candidates, and keep checking below the line.
6 min read · Updated September 24, 2026 · The LessRounds team
A pass mark is the overall score at or above which a candidate goes forward for a closer look. It’s one of the most useful settings in a screening process, and one of the easiest to get wrong, because the first number you pick is always a guess. That’s fine as long as you treat it as one. This guide covers what a pass mark should and shouldn’t do, how to pick a starting number, how to calibrate it on your first 20 candidates, and why you should keep reading below the line.
What a pass mark does, and what it shouldn’t
A pass mark sorts. It decides where you spend your reading time. It shouldn’t be the final decision about anyone, because scores are evidence, not verdicts. A short answer can hide a strong candidate, and a fluent one can hide a weak one.
In LessRounds, this is how it works:
- With a pass mark: candidates at or above it go to To review. Candidates below it are set aside in Rejected.
- Without one: candidates scoring 50 or more go to To review, and those under 50 are set aside.
- Either way: nobody is emailed about the outcome, and anyone set aside can be moved back at any time.
Screening form answers don’t count toward the score. Only the interview is scored, against the skills, levels and weights you set.
Choosing a starting number
Three things should shape your first pass mark.
How strict your scorecard is. Scores are measured against the expected level you set for each skill. If you set a high level, scores run lower, and a lower pass mark may be right. Adjust the scorecard before you adjust the line.
How many people you can read properly. Work backward from your capacity. Suppose you expect 200 completed interviews and two people who can each read about 25 candidates properly. You want roughly 50 in review, so your pass mark should end up somewhere that puts about a quarter of completions above it. You won’t know where that is until you see real scores.
Which mistake costs more. A pass mark that’s too low costs reading time. One that’s too high costs candidates, and you won’t see them, because they’re below the line. For most roles, the second mistake is worse.
If you’re unsure, start lower than feels right, somewhere between 50 and 65, and move it once you have evidence.
Calibrate on your first 20 candidates
Your first 20 completed interviews are the best evidence you’ll get about where the line belongs. Use them.
- Read all 20, including anyone set aside. Don’t skip the ones below the line. That’s where calibration errors hide.
- Judge each one before looking at the score. Read the transcript and put each candidate in one of three groups: would interview, maybe, wouldn’t interview.
- Then compare your groups with the scores. Sort by score and see where your groups fall.
- Set the line just below your lowest “would interview”. If a few maybes sit just under it, lean lower.
- Look hard at disagreements. A high score you’d reject, or a low score you’d interview, usually means the scorecard is off: a missing skill, a weight that’s too high, or a level that’s too strict. Fix that first.
Here’s an illustrative result from 20 candidates:
| Score range | Candidates | Would interview | Maybe | Wouldn’t |
|---|---|---|---|---|
| 80–100 | 4 | 4 | 0 | 0 |
| 65–79 | 7 | 3 | 3 | 1 |
| 50–64 | 6 | 1 | 1 | 4 |
| Under 50 | 3 | 0 | 0 | 3 |
Everyone you’d interview scored 50 or more, and all but one scored 65 or more. A pass mark around 60 would put 13 or so in review, catch every “would interview” except perhaps one, and set aside most of the “wouldn’t” group. The one “would interview” in the 50–64 band is worth a second look: why did they score lower than you rated them? The answer tells you what to adjust.
One practical note: in LessRounds, each candidate’s interview is frozen when they start, so if you change skills or questions during calibration, earlier candidates keep the version they took. Compare like with like.
Keep reviewing a sample below the line
Calibration isn’t a one-time job. Once the pass mark is set, keep reading a sample of candidates below it.
- Read five to ten set-aside candidates a week, focusing on those within about ten points of the line.
- Look for patterns, not one-offs. Strong candidates who gave short answers, people who misunderstood one question, or answers affected by a poor connection all show up here.
- Move good candidates back. If someone below the line deserves an interview, move them forward. Then ask why they scored low.
- Check for uneven effects. If the people falling below the line differ noticeably from those above in ways unrelated to the job, look into it. See adverse impact.
When to change the pass mark
- Too many candidates to read: raise it a few points, or check if your skills are too easy to score well on.
- Too few: lower it, or check if your expected levels are too strict.
- The role changes: a new level of seniority or a different mix of skills needs a new calibration.
- Don’t change it for one candidate. Move that candidate instead. The pass mark should reflect patterns.
- Change one thing at a time. If you change the pass mark, the weights and the questions together, you won’t know which one helped.
Keep a short log of each change: the date, the old and new number, and why. When someone asks months later why a candidate was set aside, or why the line sits where it does, you’ll have an answer that isn’t “it felt right”.
Pass marks work best alongside a clear scorecard. The scorecard decides what the score means. The pass mark decides what you do with it.