How do I interview somebody when I have never hired before?

The instinct without experience is to have a conversation and form an impression, and the problem with that approach is that it measures how comfortable somebody is in an interview. Comfort in an interview correlates poorly with performance in most roles, and it correlates strongly with confidence, which is why unstructured hiring reliably favours people who present well over people who work well.

Structure it lightly rather than formally. Four or five questions asked of everybody, in the same order, with notes written immediately afterward rather than from memory at the end of the day. That single discipline is what makes a comparison possible, and it also protects you legally, since consistent questions applied to every candidate are considerably easier to defend than a series of different conversations.

The most useful question is a real problem. Describe an actual situation from last month, give them the information you had, and ask what they would do. You are not looking for the answer you reached. You are looking for whether they ask clarifying questions, whether they consider the customer, and whether they can explain their reasoning. Somebody who asks two good questions before answering has already told you most of what you need.

Ask about something they did rather than something they would do. Tell me about a time you got something wrong and what happened next produces more information than any hypothetical, because inventing a plausible past requires more effort than inventing a plausible future. Listen for whether they take responsibility, what they learned, and whether the story has specifics in it.

Be explicit about what the job actually involves, including the parts people dislike. A first year business role is usually broader, less structured, and more interruption driven than a comparable job at a larger company. Somebody who hears that accurately and still wants it is a considerably better bet than somebody who accepts a version that was made to sound appealing, and the correction happens at three months either way.

Give a small piece of paid work where the role allows it. A few hours on something real tells you more than any interview, and paying for it is both fair and a signal about how you operate. This is the closest thing to a reliable predictor available to a small business, and it takes less time than a second interview round.

Check references properly, which almost nobody does. Ask a former manager whether they would hire the person again and listen to the hesitation rather than the answer. Ask what the person needs from a manager to do their best work, which produces genuinely useful information rather than a verdict.

Decide against a written standard rather than a feeling. Before interviewing, write down the three things somebody must be able to do and the two things you can teach. Then score against that list. The purpose is to prevent the most common small business hiring error, which is choosing the person you enjoyed talking to and discovering afterward that enjoyment was not one of the requirements.

And take your time on the first hire specifically. The cost of the wrong person is disproportionately high in a business of two, because there is nowhere for a poor fit to be absorbed and the correction consumes the founder entirely.

Tell candidates what to expect before the conversation, including roughly how long it will take and whether they should prepare anything. This costs one line in the invitation and improves the quality of what you hear, because somebody who arrived expecting a chat and received a structured assessment performs worse than they would have with notice. It also signals that the business is organised, which matters when competing for people against larger employers.