You can prepare a question list in an evening. Plenty of people do, the night before, and walk in feeling ready.
Then the candidate answers in six words and looks at you. Or talks for nine minutes without saying anything you can write down. Or describes a project in a way that does not quite add up, and you have to decide, in real time, whether that matters. None of that is in the question list.
This is the part nobody rehearses. The questions are the easy half, and they are also the half that every hiring guide covers, including ours. If you want the raw material, 50 behavioral questions sorted by competency and the interview question set for small business owners will fill your list in twenty minutes, and the complete guide to hiring your first employee puts the interview in the context of the whole process, from defining the role to making the offer.
This article is about the other half. What you actually do while the person is sitting in front of you.
The Three Jobs You Are Doing at Once
Part of why a first interview feels so awkward is that you are doing three unrelated jobs simultaneously, and nobody warns you.
You are evaluating, which requires you to stay slightly detached and take notes. You are selling, because a good candidate is deciding about you too, and in a small company they have several options. And you are hosting, keeping a conversation with a stranger from stalling while both of you are somewhat nervous.
Experienced interviewers switch between these without noticing. Beginners tend to collapse into one and drop the other two. The most common collapse is into hosting: the conversation stays pleasant, nobody is uncomfortable, and afterward you have no evidence about anything. The second most common is collapsing into selling, which produces a candidate who is enthusiastic about a role you never actually assessed them for.
You do not fix this by trying harder to do all three. You fix it by giving each one a slot. Evaluation belongs in the middle of the conversation, selling belongs at the end once you know whether you want them, and hosting belongs in the first five minutes and the last two.
When the Answer Is Nothing
Some people answer in one sentence and stop. It is rarely evasion. It is usually nerves, or someone who has been trained by previous jobs that long answers are self-indulgent.
The instinct is to fill the silence. Do not. The single most useful interviewing habit, and the hardest to learn, is letting a pause run three seconds longer than is comfortable. A large share of the most revealing material in any interview arrives in the sentence people add after they thought they had finished.
If the pause does not produce anything, do not repeat the question louder. Narrow it instead. “Tell me about a difficult project” is a broad request that invites paralysis. “Think of the last project that ran late. What was the first sign?” gives someone a specific memory to walk into, and specific memories produce specific answers.
If a candidate stays terse across several questions, say so plainly and kindly: “I want to make sure I’m not rushing you. Take a minute if you need it, and feel free to give me more detail than feels natural. I’d rather hear too much than too little.” It resets the norm without embarrassing anyone, and it often changes the rest of the interview.
When the Answer Is Everything and Nothing
The opposite problem is more dangerous, because it feels like a good interview while it is happening. The candidate is fluent, warm, and generous with words. You leave energized. Two days later you cannot recall a single concrete thing they did.
The tell is grammatical. Listen for the pronoun. “We rebuilt the onboarding flow” describes a team that included them in some capacity. “I rebuilt the onboarding flow” is a claim you can test. People rarely lie about this outright, but under interview pressure almost everyone drifts toward the comfortable “we.”
Two questions cut through it. “What was your specific piece of that?” and “What would have happened if you had been on vacation that month?” The second one is unusually effective, because it forces someone to describe their own contribution in terms of what would have been missing, and that is hard to inflate on the spot.
For the general technique of breaking a rehearsed or over-polished answer, the four follow-ups in our behavioral question bank go deeper. The short version is that the interview happens in the follow-up, not in the question.
When You Think They Are Exaggerating
At some point a candidate will describe something that does not sit right. A title that seems large for the company size, a result that seems too clean, a timeline that does not fit.
Do not confront it. Two reasons. First, you are frequently wrong, because you do not know the context of their old company, and an accusatory turn ends the useful part of the interview. Second, even when you are right, a confrontation gives them a chance to construct a better story in real time.
Get more detail instead. Exaggeration collapses under specifics, and it collapses on its own, which spares everyone the confrontation. Ask about the parts nobody rehearses: who disagreed with the approach, what the first version got wrong, what they would do differently. A person describing something they genuinely did has an inexhaustible supply of texture. A person describing something they watched happen runs out in about ninety seconds, and you will hear it.
Then separate two different things in your own head. Inflation is not the same as invention. Someone rounding their role up from contributor to co-lead is behaving like most candidates under pressure. Someone describing a project they were not on is a different matter entirely. The first is normal and you correct for it. The second is disqualifying, and it is what reference checks exist to catch.
When They Know More Than You
If you are hiring your first specialist, you will interview people who are better at the work than you are. This is the situation that makes new managers most uncomfortable, and it produces two bad reactions: pretending to follow, or over-correcting into skepticism about someone whose expertise you cannot evaluate.
Say it directly: “I’m not going to be able to assess the technical depth here, so I’m going to ask you to explain the reasoning instead.” Good candidates respect this. It also happens to produce better signal, because you have moved from testing knowledge, which you cannot judge, to testing judgment, which you can.
Judgment questions work on any expertise, including expertise you do not share:
- “What is a common approach in your field that you think is wrong?”
- “Tell me about a tradeoff you made where the other option was defensible.”
- “What would you need from me to do this well, and what would make it impossible?”
You are listening for whether they can hold two sides of a decision, whether they know why they chose what they chose, and whether they can explain it to someone outside their specialty. That last one matters more than it looks, because you are going to be that someone for as long as they work for you.
If your remaining doubt is genuinely technical, do not try to resolve it in conversation. Bring in one outside person for thirty minutes, or use a small paid work sample.
When They Ask You Something You Cannot Answer
They will ask about runway, growth plans, why the last person left, or what happens if the business has a bad quarter. First-time interviewers tend to improvise something reassuring.
Improvising is the worst option available, because you are establishing, in the first hour of the relationship, how you behave when you do not know something. If they take the job, they will remember. If your answer turns out to be wrong, they will remember that too.
“I don’t know” is fine. “I don’t know, here is how I’d find out, and I’ll email you tomorrow” is better, and following through on it is one of the strongest signals you can send a candidate about what working for you is like. Very few interviewers do it.
On the uncomfortable questions specifically, honesty outperforms polish. If the last person left because the role was badly defined, say that the role was badly defined and describe what you changed. A candidate who hears an honest account of a real problem learns that you will tell them the truth. A candidate who hears a smooth non-answer learns the opposite, and the good ones are listening for exactly this.
They Are Interviewing You, and in a Small Business That Decides It
For a candidate choosing between a small company and a large one, the manager is a much bigger part of the decision than the job description, because there is no HR layer, no internal transfer, and no second manager to move to if it goes badly. They are trying to work out what you will be like on a difficult Tuesday.
This is why the interview is not only an assessment. Three things carry disproportionate weight, and all three are free:
Whether you ran the meeting well. Starting on time, knowing their background without rereading it in front of them, and finishing when you said you would are all read as evidence of how you will manage.
Whether you were specific about the role. Vagueness in an interview reads as vagueness to come. Being able to say what the first ninety days look like, what success means, and what the hardest part of the job is tells them you have thought about the work they will actually do.
Whether you told them the bad parts. Every job has some. Naming two of them makes everything else you said more credible, and it filters out people who would have quit over them in month three anyway.
Save this for the last ten minutes. Selling before you have assessed contaminates your read, because it is difficult to evaluate someone honestly after you have spent half an hour persuading them.
Your Nerves Do More Damage Than Your Inexperience
Being new is not the problem. Candidates do not expect a first-time interviewer to be smooth, and most of them are more nervous than you are.
The problem is what nervousness makes you do, and it is predictable enough to watch for:
- You talk too much. The rough benchmark is that the candidate should be speaking for most of the conversation. If you leave with a hoarse voice, you conducted a presentation.
- You lead the witness. “So you’re comfortable with ambiguity, right?” is not a question, it is an instruction. Ask what they did, not whether they agree with a description of themselves.
- You agree with everything. Nodding along to an answer you found unconvincing feels polite in the moment and leaves you with unusable notes.
- You decide in the first four minutes and spend the rest confirming it. This is the expensive one, because it feels like clarity. If you catch yourself already sure, deliberately go looking for the opposite evidence before the interview ends.
A structure helps more than confidence does. Same questions, same order, notes written down during rather than after. That is also the entire argument for structured interviewing: it is not that it feels better, because it does not, it is that it removes the parts where your nerves and your memory make things up.
Taking Notes Without Losing the Room
Write during the interview, not after. Memory does not store an interview, it reconstructs one, and by evening the reconstruction is heavily colored by whether you liked the person.
Say at the start: “I’m going to take notes while we talk so I can be fair to everyone I’m speaking with.” Nobody minds once it is named, and it stops the pauses from reading as disapproval.
Write quotes rather than judgments. “Rebuilt onboarding, cut two weeks, disagreed with her manager about scope” is usable next week. “Good energy, strong communicator” is a feeling you had on Tuesday and will not be able to defend against a different feeling you have on Thursday.
Leave five minutes between interviews to finish your notes. Back-to-back scheduling is the most common reason good notes never get written, and unwritten notes are the reason strong candidates lose to whoever happened to interview last.
What This Guide Leaves to Other Pages, on Purpose
The interview is one step in a longer process, and repeating the rest here would make this page worse rather than more complete. Where the detail actually lives:
- The questions themselves. 50 behavioral questions by competency for a competency-first bank, or interview questions for small business owners for a set organized by what you are trying to learn, plus the timed interview agenda, red flags, the trial project, and the scoring system.
- What you are legally not allowed to ask. Covered in both articles above, and worth reading before your first interview rather than after.
- Everything around the interview. Defining the role, writing the job post, making the offer, and onboarding are all in the complete guide to hiring your first employee.
- The step straight after this one. Inflation is normal and invention is disqualifying, and the difference is what reference calls exist to catch. Reference check questions covers who to call, when to call them, and the fifteen questions that resist a reflex compliment.
- Whether you should be hiring at all yet. If the role, the budget, or the decision process is still fuzzy, the Am I Ready to Make My First Hire? self-assessment is a faster way to find out than discovering it during interviews.
The Standard to Hold Yourself To
You will not run a great interview the first time. The realistic goal is narrower and much more useful: leave the conversation with written evidence about specific things the person has actually done, and with the candidate still interested in the job.
That is it. Not certainty, which is not available at this stage from anyone. Evidence and interest.
The cost of getting this wrong is worth knowing before you start, because it is the number most first-time employers underestimate by a wide margin. The cost of a bad hire calculator works through salary, lost productivity, your own time, and the cost of rehiring. Most people are surprised, and it is a useful thing to have in mind the next time you are tempted to skip the follow-up question because the conversation was going pleasantly.