In my last article, inspired by Indi Young's book Practical Empathy, I proposed that we might expand the thinking and effectiveness of our organizations by developing empathy during a technical interview, through what Indi calls a dedicated listening session. In this article, I will explore what this technical interview might look like.
Why Does It Matter?
First, it is worth revisiting why this matters. Much has been written about how technical interviewing is ineffective. We tend to filter out people based on algorithms and data structures trivia, or "Aha!" style brain teasers. Just because a person can remember a bunch of facts, or has heard your brain teaser question before, has little or no bearing on this person being an effective addition to you organization.
Dan Pink, in his book Drive, describes how people tend to perform worse under pressure and when incentives are large. Their focus narrows and they fail to see other solutions. They can develop a functional fixedness, where they focus on using something only in a conventional way, failing to see other solutions. For most people, an interview is a stressful situation and the incentives can be large, perhaps even career changing. Consider the affect this has on the candidate's answers to algorithms or brain teaser questions. Dan also details how mastery is hard work and that perseverance and a passion for long-term goals are the best indicators of success. How are we evaluating this in technical interviews?
After filtering candidates based on these poor predictors of success, we subject the candidates to a a few conversational interviews to assess "cultural fit", yet another poor predictor of success. If you have never worked with someone like this person before, how can you know if they are the "right fit" or not? Candidates in these interviews can often give you answers they think you want to hear, rather than expressing themselves genuinely. Lauren Rivera has researched how these unstructured interviews lead to biases based on the interviewer's personal enjoyment of fit, rather than aligning with organization values. She notes that interviewers often make up their mind within the first few minutes, almost exclusively based on their subjective perceptions. She concludes that fit has become a catchall used to justify hiring people who are similar to decision makers, rejecting people who are not. Extroverts seek extroverts. Introverts seek introverts. Open-source contributors hire open-source contributors. We are attracted to individuals with a similar pedigree to our own. I'm even more cynical and will suggest that, in addition, many people will also consider hiring people they believe they can control to meet their own needs.
A lot of attention has been paid to diversity in the workplace based on observable differences like gender, race, and age, but diversity of working styles and ways of thinking is an even more latent issue. We want diverse ideas when we are trying to innovate, or when working on tasks that involve creativity and complex decisions. We want different perspectives and new ideas. We need to appreciate the differences in others, rather than just hiring copies of ourselves. Please Understand Me is a wonderful book for understanding how you think and how others think differently from you. As Indi describes in her book, observation alone is not enough. Correlation is not causation. We need to understand people at a deeper level. This involves listening and developing empathy. Let's explore what that might look like in a technical interview.
An Interview as a Listening Session
What might an interview designed as a dedicated listening session for developing empathy for a candidate look like? First, it should be conducted one-on-one and not combined with any other aspect of the interview, either technical or non-technical. Even introducing disarming topics like the benefits package or your touted "work-life balance" will have the affect of controlling the conversation and influencing how this person responds to you. Second, the conversation should remain largely technical, after all, you are trying to evaluate this person for a technical role within your organization.
What would be some good topics? Questions like the following might be good places to start:
- "What is the most difficult problem you've ever solved?"
- "What is your favourite programming language?"
- "What makes you feel most creative?"
- "What is the best team you've ever been a part of?"
- "What rules of thumb do you use when debugging a problem?"
Then just let the candidate guide the conversation and dig deeper into this person's thinking, preferences, reasoning, and experiences, using the techniques Indi describes. Pay particular attention to setting aside your own reactions and judgments and appreciate the other person's thinking.
I think code review could be incorporated into a listening session, especially if your interview process requires submitting a programming assignment—reviewing the candidate's own work would be ideal—but be sure to stay neutral and curious when discussing the code, never implying the candidate is wrong, or providing your own solutions. I think the candidate will generally enjoy this interview and it will leave a strong impression, since you are focusing exclusively on them and developing a real understanding for them and their work.
Something interesting to consider: at the end of the listening session, should you take questions from the candidate? I'm going to say, yes, for two reasons. One, the candidate's questions will serve as an opportunity to not just answer the question at hand, but to find out why the question is important to the candidate. Second, if you develop a connection with the candidate, the candidate may ask you questions that they will not ask other interviewers. This may be a very important opportunity for them to decide if they want to work with you and for you to construct a deeper understanding of the candidate. I think questions, however, should only be taken at the end of the listening session, not interjected throughout.
After listening to the candidate and developing empathy for the candidate, you ultimately need to make a hiring decision. This is somewhat juxtaposed to the listening session where you were trying to reserve judgment, but hopefully the greater understanding you have developed for the candidate helps you make a more enlightened decision. One might also argue that this interviewer should not pass judgment on the candidate at all, but, rather, just let the empathy they have developed for the candidate inform the discussions that lead to a hiring decision, giving insight into the candidate's thinking and motivations.
Not everyone will be suited to conducting such an interview. It will take interest, aptitude, and practice on the part of the interviewer to become skilled at this interview. Interviewers from Human Resources might traditionally have more skills in this area, however, I think the person conducting the interview needs to be technical in order to keep the conversation highly technical. It will be important for this person to understand that other people have different ways of thinking and solving problems than they might. As I already mentioned, the book Please Understand Me is a good place to start building an understanding for how you think and how others may think differently. In preparation for the interview, you might also consider incorporating a personality test, like the Myers-Briggs Type Indicator, to gain perspective on how the candidate thinks.
Finally, when should such an interview be conducted? I think it makes sense to do it early on in the process, perhaps as a phone screen, before filtering out candidates indiscriminately based on not being able to reproduce a merge sort or what have you. Indi actually prefers to conduct listening sessions over the phone, because it is easier to focus on listening and not have facial expression or body language influence the conversation. I think there is probably also value in repeating a listening session later on in an in-person interview, to develop another perspective.