The onus is on the interviewer

This morning my wife and I were grabbing coffee at a local coffee shop. As I was waiting for our coffee near the bar and I overheard a man interviewing a woman. “So, uh, is this a job you see yourself doing for the long haul…?” he asked. Naturally, her response was an emphatic “Yes, of course!” I mean, why would she say anything else? If she had said “I’m going to give this job a year and then re-evaluate my options”, would he even consider hiring her? I really started to wonder what the point of that question even was - it was a leading question that gets one answer unless the candidate is brutally honest (which is really risky if you need a job) or they don’t really care at all about getting this job. Sadly, these types of single answer, shallow, questions are more the norm rather than the exception. I’ve given hundreds of interviews throughout my career in tech and vary rarely do I find an interviewer that is completely prepared to get what they need from a candidate (myself included).

Now, interviewing is hard. It’s one of those skills that really befuddles people - probably because an interview is completely skewed by the perception of the interviewer. Perception is, by definition, subjective so you can’t guarantee a consistent experience. Interviewing is qualitative and really hard to measure. Measuring anything well requires a lot of samples so outside of the major tech companies, most don’t interview enough to even measure anything. Even within big tech, rarely does a single interviewer actually perform enough interviews to show statistical significance between the interviewer and candidate success. And finally, even if there is a single interviewer with enough interviews to be measured, odds are the interviews happened over a long time period - long enough that the interviewer has changed perspective, learned to be more (or less) effective, or changed their opinion about what a good candidate looks like to the company.

I believe that only good way to interview is to think long and hard about the characteristics you want in a candidate and then craft the interview experience. These characteristics probably need to be foundational to the company culture and not a checklist of desired skills. For example, if the company values employees that are inquisitive, and like to learn for the sake of learning, you can craft prompts to try and determine if the candidate actually cares about learning.

“Tell me about something you recently taught yourself.”

It’s innocuous, but if the candidate can’t speak to this, they either don’t prioritize learning or they are trying to find something you’ll like to hear. What’s important here, though, is that you - the interviewer - are prepared to dive in and find out exactly what the person learned, why they learned it, and how they learned it. If a person really loves learning, you will be able to tell. As you dive in to find more details the candidate will light up. They may come up with other examples on the fly. They may keep talking about what they learned with enthusiasm.

Say you are interviewing for an engineering role. You ask the candidate to tell you something they recently taught themselves and the candidate answers with:

“I recently taught myself how to properly cook a steak.”

If you are a bad interviewer you will dismiss this answer, probably by asking for something “tech related”. If you are a good interviewer you will probe for deeper answers. Why steak, and why not cooking in general? How did you teach yourself to cook that steak? What does it mean for a steak to be cooked properly? What did you learn about cooking steak that was new or interesting to you? And on, and on, and on… You can really keep going if the person actually cares about learning to cook steak. The problem here is that it’s on you - the interviewer - to go deeper. You need to do all of the work. The candidate should be able to effortlessly explain what they did in the ideal case.

Let’s try a different characteristic - trust. Trust is probably the most important criteria for a successful relationship. If you can’t trust someone they don’t deserve a place in your life or your business, period (I’ll dive into this more in a future post). You won’t be able to identify a candidate’s trustworthiness from a single prompt or question, so again, you need to dive really deep into the answers they give.

“Tell me about a work relationship that you have (or had) where you felt mutual trust between you and someone else.”

Some follow up questions here could be:

• How did you know there was mutual trust?
• What did you do to establish trust with this person?
• Who was this person? A manager? A peer? A report?

If there really was trust it should be obvious as the candidate explains the relationship to you. Think about the high trust relationships you have - they are obvious. The people you trust are confidants, you can probably tell them effectively anything. The people you trust won’t micromanage you and you won’t micromanage them. The people you trust come through for you and you come through for them. This trust didn’t come for free, either. You probably have tenured relationships with the people you trust and you can probably explain what has happened over time to build that trust. Trust is intuitive - you can feel when it’s there or when it’s not. Using your intuition to “feel” the trust in the candidate’s relationship as they explain it to you is okay. Remember to ask more questions if you’re not quite sure what’s being said.

Over time you can build a model for generating useful interview prompts. Start with the characteristics you want or need in an employee, for example, someone with the inclination to ask for clarification when expectations are ambiguous, someone that you can trust to come through for you, someone that takes pride and ownership in their work, etc. From here, generate the necessary questions or prompts to spark conversation. I call them “prompts” simply because it’s easier to generate a more open ended conversation than when I create questions, but it’s the dealer’s choice. Finally, dive in deep. Ask many, many, follow up questions. A single prompt could easily spawn a 30 minute conversation - you are trying to understand the candidate’s core values, anyway. Core values don’t emerge from short answers to simple questions. Ultimately, you want to avoid prompts that candidate’s can game by giving easy answers that tell you nothing (such as a closed ended question with a single, obvious, answer like the example I used in the beginning of this post).

I realize that I’ve only touched on behavioral questions. I plan to address some other facets of the interview process in follow up posts, like how to devise great technical questions, or how to build a great interview experience. The biggest problem is that interviewers tend to be unprepared. They don’t really know what they want or what they need falling back to a naive and trivial list of “skills”. Skills can be learned over time, but character isn’t something that changes very fast - if at all. Identify the candidates with the ideal character, then work on the skills if there are noticeable gaps. If you do a good job hiring training won’t be much of a problem, anyway.