How Many Interview Rounds Do You Actually Need?

You've just watched a good candidate go quiet after round four. They were warm through rounds one and two, still engaged in round three, and then... nothing. Another company moved faster and they took it. You didn't lose them to a better offer. You lost them to your own calendar.

Most small companies don't design their interview process. It grows by accident — someone adds "maybe get the CEO to meet them" for one hire, and three years later every hire meets the CEO, whether the role needs it or not. Nobody ever removes a round. They only get added.

So how many is actually right?

The short answer

For most roles at a 10–200 person company: two to three rounds. A screen, a real conversation with the hiring manager, and — for anything technical or high-stakes — one more step that tests the actual work. Four is defensible for senior or leadership roles where you're also checking fit with multiple people they'll work closely with. Five or more is almost never about better decisions. It's about nobody wanting to be the one who says yes.

What each round should be for

The problem isn't the number. It's that most rounds ask the same questions in a different room. If round two just repeats round one with a different interviewer, you haven't learned anything new — you've just made the candidate say "tell me about yourself" twice.

Each round should have one job:

1. The screen (20–30 min). Rule out, don't rule in. Can they do this job, roughly, at this salary, starting roughly when you need them? That's it. Don't go deep here.
2. The real conversation (45–60 min). This is where the hiring manager digs into specific past situations — what they actually did, not what they'd do in theory. This is the round that predicts performance, so give it the most time and the sharpest questions.
3. The proof round. A work sample, a structured test, a case relevant to the actual job. This is the round most companies skip and most regret skipping. Someone who interviews beautifully and someone who does the work well are not always the same person, and no amount of conversation tells you which one you've got. This is also the round most people cut first when they're in a hurry — which is backwards, because it's the one that actually changes your decision.
4. The fit check (optional, 30 min). A future teammate or a second senior person, mainly to catch something the hiring manager might have missed and to let the candidate ask questions from someone who isn't selling them the job.

If you're running more than four rounds, ask what round five is actually testing that round two didn't. Usually the honest answer is "reassurance," not information.

Sizing it to the role

• Entry-level or high-volume roles: one screen, one conversation, done. If the role is genuinely entry-level, a long process filters out exactly the candidates you want — people with other options who won't wait three weeks for a warehouse job.
• Individual contributor, skilled: screen, hiring manager conversation, work sample or test. Three steps, nothing more.
• Manager or lead: add one round with a peer or direct report, because how they treat people who aren't deciding their fate tells you something the hiring manager conversation can't.
• Senior leadership: four rounds is reasonable here, because the cost of getting it wrong is much higher and more people need to have actually met the person. Just make sure each one has a distinct purpose, written down before you start, not decided on the fly.

The real cost of too many rounds

Every extra round doesn't just cost you time. It costs you candidates. Good people — the ones you actually want — usually have other conversations going. A process that takes five rounds and three weeks tells them something about how decisions get made here, and it's not flattering. The candidates who stick around for round five are sometimes the ones with the fewest other options, which is exactly the opposite of what you want your funnel to select for.

There's also a quieter cost: interviewer fatigue. If six people each spend an hour on every candidate, you're burning a full working day of company time per hire before you've made an offer. Nobody notices this cost because it's spread across six calendars instead of one line item, but it's real, and it's usually the first thing worth cutting when hiring feels like it's eating your week.

What to cut first, if you need to move faster

Start with any round that exists mainly to make someone feel included, not to gather new information. Then look for duplicate rounds — two people asking variations of the same behavioural questions. Combine them, or split the questions so each interviewer owns a different angle and compares notes afterward instead of covering the same ground twice.

What you shouldn't cut is the proof round — the one where you see actual work, not a description of past work. If you're short on time, this is often better run before the final conversation rather than after it: a short, scored skills test or work sample that the candidate does on their own time, so the in-person rounds you do run are spent on judgement calls a test can't make, not on re-verifying whether they can do the job at all. AssessFit is one way to do this without adding a scheduling headache — the test runs asynchronously and it's free for candidates either way.

The judgement call nobody tells you about

There's no universal right answer here, and be suspicious of anyone who says there is. A five-person founding team hiring their first ops lead should probably talk to that person more than a 150-person company hiring their tenth support rep. The honest test isn't "how many rounds is standard" — it's whether you could explain to the candidate, before they start, exactly what each round is for and what happens after it. If you can't answer that for one of your rounds, that's the one to cut.