The phone screen: a 15-minute guide that saves you hours of interviews

The screen is a filter, not an interview. Its only job is to decide, quickly and cheaply, whether a candidate is worth a real interview slot, that's it. So you're not assessing fit in depth; you're checking for the few deal-breakers that would waste everyone's time if discovered an hour into a panel. Keep it to 15 minutes, keep it friendly, and resist the urge to turn it into a mini-interview. The whole value is in protecting your team's expensive hours from obvious mismatches.

Most screens come down to four checks. Motivation: why this role, why now, do they even know what we do? Must-haves: the one or two non-negotiable requirements, confirm them out loud rather than assuming the CV. Money: are their salary expectations in the same universe as the budget? Finding a gap here after three rounds is a classic, avoidable waste. Timing: notice period, other processes they're in, when they could start. Four honest answers tell you whether to invest a real interview or to send a prompt, kind no.

Ask every candidate for a role the same core questions, in the same way. This isn't bureaucracy, it's what makes the screen fair and the answers comparable, and in Germany it's part of running an AGG-compliant, bias-resistant process. A consistent screen also protects you: when two candidates are close, you're comparing like with like instead of half-remembered impressions from differently-run calls. Have the questions written down in front of you, and note the answers as you go.

The screen is worthless if its outcome lives only in your memory until the next time you open the candidate. Right after the call, record the verdict and the why, advance them to the interview stage or send the rejection, while it's fresh. In KI BMS the call notes and the stage move sit on the application itself, so the interviewer who picks them up next sees exactly what the screen surfaced, and nothing has to be re-discovered. A screen that isn't written down just gets re-done badly later.