Developing effective interview questions

I read a blog post a few years ago that really influenced how I thought about the questions we ask in Happiness Engineer interviews. In the past our interview had centered on questions about a person’s background with WordPress along with their thoughts on what makes good customer support. These were often interesting conversations, but we weren’t getting as much of a sense of fit as we would have liked. This resulted in more candidates being advanced to the sample questions assignment or even to trial before we saw signs that they weren’t going to be successful. I’ve thought back on that post from time to time but have not been able to find it again (dangit!). The memorable takeaways for me were:

  1. Avoid generic questions like, “What are your strengths and weaknesses?” Have a specific reason for each question you ask, and an idea of how the response will inform your evaluation of the person’s fit for the role.
  2. Ask a question that probes what kind of research they did prior to the interview (as a sign of their genuine interest in the role and the company).
  3. Be attentive to the candidate’s ability to recall information discussed earlier (as a sign of their ability to observe and retain relevant details).

We discussed things on our team and updated our question list to incorporate the three items above. We started this process several years ago, but we’ve continued to tweak our phrasing over time. I thought it could be useful to other talent professionals to share our evolution on this.

Item 1: Have a specific reason (and an ideal answer) for each question you ask

Evaluating how they think about support
Original:Do you think the customer is always right?
Current:How do you approach it when your best efforts to help a person aren’t working, whether through a misunderstanding or lack of experience on their end?

The answers we would get to our original version of the question, Do you think the customer is always right? didn’t reveal a lot about fit for the role. Most people responded with general assurance that they would try to make a customer happy whether they were right or wrong. A much more useful question for us turned out to be: How do you approach it when your best efforts to help a person aren’t working, whether through a misunderstanding or lack of experience on their end?  

Great: “I would provide numbered steps and screenshots, and ask questions until I understood which step they were stuck on so that I could explain that step more clearly.”

This answer indicates familiarity with support strategies and flexibility to try different approaches to find what works.

Lower fit: “I would keep trying until they get it.”

The lack of specifics in this response suggests lower aptitude for adapting their approach to fit individual users.

Red flag: “I would see if someone else can help them.”

While consulting with colleagues is a valid strategy, with no other context this response suggests a lack of initiative for independent problem solving.

Evaluating VALUES fit
Original:Can you tell me about something you’ve achieved in your life that you’re proud of?
Current:Can you tell me about something new you learned recently?

Our intent with asking about an achievement had been to get a sense of a person’s initiative outside of a structured work or school role. The question had been inspired by thinking about the accomplishments of some of our high-performing Happiness Engineers, like authoring a book or running a business or training for a marathon. These sort of accomplishments are a signal that a candidate has the internal motivation to be successful in a role that lacks the typical external structure found in a more traditional workplace. However in practice the answers we got were often not very useful to us in an evaluative sense, and even sometimes led into topics that would best be avoided in an interview, like: “I’m really proud of my kids.” Our updated question gets at the same general idea (are they self-motivated) but dials in a little tighter on their culture fit: Can you tell me about something new you learned recently? Either something you learned for work, or something outside of work is fine too.  

Great: “I’ve been learning Spanish on Duolingo.”

This suggests a good fit for our always-learning culture.

Lower fit: “I learned that tongue prints are as unique as fingerprints!”

Responding with trivia rather than a skill (particularly in the context of a job interview) is missing the point a bit. We tend to find that people who interpret the question this way have weaker responses to other questions in the interview also.

Red flag: “I don’t really have the time to take a class or anything like that.”

This response suggests low awareness of online resources and/or low interest in personal/professional growth. Not a close alignment for a fast-moving tech company where tools and processes can change frequently.

Item 2: Ask a question that probes for their genuine interest in the role and in Automattic

Those of us on the hiring team approached our own applications to Automattic by devouring every bit of online information we could find about the company and the role. Early on we had an assumption that most applicants we interviewed were similarly informed, but it became clear that some candidates had lower information than we had assumed. Thus, questions about the company and the role are a good way to evaluate what preparation the candidate has done prior to the interview (a strong signal of their genuine interest level).

Evaluating their interest in the company
Original:Do you know how our hiring process works?
Current:Our hiring process at Automattic is a little unusual. What do you know about how it works?

We started out with a simpler version of this question: “Do you know how our hiring process works?” We were expecting answers like: “Yes, I know there’s an interview and a trial project and then a chat with the CEO.” But the answers we received tended to be more brief and vague, like: “I think so!” We realized we needed to make it less yes/no and more open-ended. We tweaked the phrasing on this question until we got to a version that produces a meaningful response much more consistently: “Our hiring process at Automattic is a little unusual. What do you know about how it works?”

Great: “I’ve read there’s an interview followed by sample questions, a paid trial project, and a chat with the CEO.”

This clear and concise answer demonstrates they’ve done research on Automattic.

Lower fit (but often ok): “I think I read something about a trial project, is that right?”

People may answer hesitantly for fear of being mistaken on a detail and getting the answer “wrong,” so we view this in context with their other responses.

Red flag: “Not much! Can you tell me about it?”

A response like this one often correlates with giving other low-information responses over the interview.

Evaluating their understanding of the role
Current:Can you tell me what kind of work you would expect to do in a typical day as a Happiness Engineer?

We added this question to probe for the candidate’s understanding of what the role is (and isn’t). Some folks who apply for Happiness Engineer focus on the “engineer” part of the job title and miss the “support” part of the job description. They assume that it’s a junior developer role, or that it will be like working for an agency and they will be assigned website projects for clients. Besides the formal job description on, there are numerous blog posts by current staff describing what HEs work on in great detail. Strong candidates come to the interview having done that research and having already gained a clear understanding of the role. We check for this by asking the candidate to tell us what they would expect to work on.  

Great: “I would answer questions over live chat and email, report bugs, and contribute to support docs.”

This clear and concise answer demonstrates they understand what an HE works on.

Lower fit: ”I would be an active member of a global team that provides 24/7 support via live chat, tickets, forums, and screenshare sessions to help people use Automattic’s products, including, WooCommerce, Jetpack, and more.”

This isn’t necessarily a wrong answer, but we take it in context of their overall responses. It’s usually better if they answer the question in their own words without resorting to copypasta from our Careers page.

Red flag: “I was hoping you would tell me!”

This response indicates low understanding or recall of the work described on the HE job page.

Item 3: Watch for whether they retain details previously discussed

This is challenging to design a question around in a text-based interview, where the candidate can simply scroll up to see what was said earlier. However we do sometimes see signs of how they have processed the prior discussion in the questions they choose to ask us at the end of the interview. Some candidates ask generic questions that could apply to any job/company and don’t relate back to anything already discussed, or anything specific to the position. Some candidates will ask questions on topics that were already covered, or that indicate they didn’t really take in the conversation about what the role is and isn’t. This suggests that they came prepared with only one question, and weren’t able to adapt on the fly or deviate from their planned script.

Evaluating their closing questions in context of the prior discussion
Current:That’s about all I had to cover for now. Do you have any questions for me? Anything I can clarify about the position or about working at Automattic?

Great: “Can you tell me how the work is organized? How does an individual Happiness Engineer know what area is a priority to focus on when they start work each day?”

This question indicates they’ve given thought to the job description of “providing support via live chat, tickets, and forums” and realized there would likely be systems in place for organizing the work; they’re curious about what those systems are.

Lower fit: “Who does the position report to?”

This question is copied from a “good questions to ask” list; it’s a bit of checking the box just to have a question, and a missed opportunity to ask something more meaningful about the position responsibilities or the company culture. There’s no connection to anything already discussed during the interview. It’s a pretty generic thing to ask, especially if they don’t have any other questions.

Red flag: “What would I work on in a typical day?”

This question is concerning because it’s so close to the question we already asked them earlier (“What would you expect to work on in a typical day”), and to which we always follow up with a clear description of the typical daily tasks. It’s not a good sign if 15 minutes later they ask us to describe the work again, without any other context.

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s