Culture Fit Interview Questions

The purpose of the interview for the interviewer: deciding between "no, don't hire this person", "maybe, we'll decide later with more data or after comparing with another person", and "yes, hire".

The purpose of the interview for the interviewee: deciding "yes/no, I want to continue with this company" and conditionally if yes trying their best to get the interviewer into the "yes, hire" state.

I'm only going to pick on one of these questions, but they all have the same problem in that they're not (to me) very effective means at fulfilling the purpose since they'll select for candidates most capable of reading your mind, not actually really useful candidates. When you ask "what's your take on accessibility on the web?", the candidate is thinking many things at once. Here are some: 1) do I have a take and what is it? 2) what's the answer the interviewer wants to hear? 3) if my answer is wrong does that make it impossible to get into the "no, don't hire" state? 4) will I know once I answer?

Maybe you want a take that says accessibility is important despite the added costs, because humanism or whatever. If they say accessibility isn't important, because the costs don't justify it, you put them in the "no" bucket. Or perhaps it's vice versa, or you analyze the issue through some other framework. Maybe you really just want to see if they can converse about it at all, and will devil's-advocate the opposite of what they say, and only put them in the "no" bucket if they can't converse or start screaming at you.

In any case, I don't think you're being fair to the candidate and you're likely wasting time. If you just want to test ability to converse, and don't actually care what they personally believe, state that in the question and don't ask for their actual belief: "I'd like to have a sort of philosophical discussion with you about accessibility on the web. Let's imagine I ask you ... and you feel ..." If you actually have a specific range of answers in mind that can put someone in the "no" bucket, put that information in the job description requirements. "Expected to design with accessibility in mind." fits the first of my maybes, "Expected to move fast and not spend time on non-MVP work like accessibility." fits the vice-versa. Presto, no mind games, no wasted time for everyone because it wasn't clear until you asked your question that you hold opposite views and thus this is a "no". Maybe you're worried about liars, which you have to be anyway since plenty of people apply to coding positions without being able to code, so you might ask a more specific question (like we do by asking them to code something) around accessibility that makes clear what conclusion you expect (it matches the job description) and that you're looking for some sort of reasoning for why that conclusion is such in their mind.

Maybe since you're mentioning "end up being really useful" (as opposed to just useful) you don't weigh answers to these questions as hard "yes" or "no" filters, but just "maybes" that you can subjectively reflect upon later (e.g. by adding up a bunch of "maybes" you've recast to point-weighted soft yes/nos that can cancel each other). Fine, you can still put "Bonus:" in the job description, rather than "Required:", so that candidates know ahead of time that if they can only get to the "maybe" state the presence/absence of those certain "Bonus" attributes will influence their chances of moving from "maybe" to "yes". If they're already uncertain about "yes", and see a lack of "bonus" attributes on top of that, they're likely to not bother, again saving everyone's time.

It's fine to distinguish between "really useful" and "useful". Questions that can distinguish between degrees of "maybe" between candidates aren't bad, but they should be back-loaded as much as possible, and only used when the front-loaded yes/no questions have been asked and you're still in a state of uncertainty about which candidate would really be better, lamenting that you only have the budget for one of them. How many of those do you get?

- What's your mechanism for bias self-check?

Debate. I like debating, I have to admit. It's a great way to sharpen one's logic. But, one has to careful to not get labeled as "argumentative".

- If someone gives you specs and you notice that something is off, what do you do?

I always have suggestions or questions on any non-trivial spec. I list them up and email them out to the project manager. I do try to be polite in my criticism, though. Example: "I'm concerned that X may confuse users. Here's an alternative to consider."

- If you have to solve a problem you haven't solved before, how do you approach it?

I try to ponder such longer than normal rather than go with the first approach that pops into my head. I may work on something else while the idea dances around in the back of my head. Sometimes I dream up solutions at night. It sounds cliche, but it's true.

- What's your take on accessibility on the web?

Managers and users often like fancy UI's that may run into accessibility problems. It's a tricky thing to balance. Eye-candy does "sell" a UI, and accessibility often hampers that. Sometimes I try to find ways to spruce up a UI that don't harm accessibility, such as improving logos or including interesting "side" graphics that don't interfere with the primary function.

- What's your process like for deciding that you're at the point in your career where you can mentor others?

If I feel I can help somebody without offending them, I will often just jump in and do task-specific mentoring. But I wouldn't want to do such full time.

- What do you prefer to do when you see someone else getting nit-picked?

I will defend them if their ideas or work have merit. If not, there's probably a reason they create team friction, and to be frank, they probably should change careers into something that's a better fit for them. Often times such people are not "bad", rather just in the wrong field.

- You're just about to finish a feature and have a great idea for improving it. What do you do?

I'll check with my supervisor. If I'm enthusiastic about the better variation, he or she will typically allow me to go with the better one once they know me. If time is tight, it may have to wait until the next version. I will work extra hours if I feel it's a compelling feature.

My take is that you need to project the image of a professional, match your resume experience, don't be combative and basically show that you can be a valuable drone. Even the fact that you can bullshit this with a straight face and not go candid on the interviewer is a positive data point, as that's basically 80% of working in a corporation. But again, the hiring manager can cherry pick positives/negatives based on what he actually wants to do.

It’s important to understand that these questions are just the starting point of a conversation to understand how a person thinks and acts. It’s impossible to make things up with a good recruiter because there’s no way you know beforehand how deep and where he will go with his follow-up questions

Noise might not contain information but the presence or absence of noise does.

The target audience for this is small to medium businesses, not corporations or competitive startups, who would be potential customers. It's more of a starting place for people who do not think about this much.

My guess is that HN is not the target audience for this blog post, so take what the simplicity of what is said with a huge grain of salt.

That's so funny, I read these question's and though I could consolidate them into a single question:

"How well can you bullshit and talk about things entirely immaterial to work?"

Seems a lot of people are chiming in to say how great question #3 is (how do/did you solve a problem at work)...I've gotten this in interviews and told them the Mad Men/Don Draper story of needing a new ad for my client "Lucky Strike" because the attorney General was going to prohibit our existing ads, complete with the pitch how I convinced the client its the greatest opportunity in marketing, asked them how they make their product....stopped them in their tracks, said that's it: "Its Toasted". Interviewers seem to love that story, and if they have seen Mad Men they seem to love it even more. Other times I told them about selling Mangos when I was a child...they love that one too, nothing more bullshitters love than other bullshitters.

You might find it telling that Google is censoring word "bullshit" on YT as of yesterday : https://www.youtube.com/watch?v=bHvkEPL7mVA

People with a reputation for getting results get to skip this step, the bottom line matters more than your character. But any company will hold up a boundary at the door to make sure you fit the mould, or at least don't sink the ship.

The fundamental problem of bringing the unknown into the known has to have some boundary, what that is that makes sense usually just rides a trend wave. Character and behavioural stuff is all the rage right now. It'll pass onto other things soon enough.

1.) In terms of a daily schedule, on a scale from 1-10 how much are you a creature of habit?

2.) If you need a quick answer from a person 20 ft. away, do you send an email or walk over?

3.) What strategies do you have for completing tasks you don't like doing?

From my perspective interviews can be super biased.

Skill set aside, I think we naturally want to hire someone who answers our questions in the same way we would.

IMHO.

And it only gets easier after your first gig.

In that instance, where an engineer can easily leave your company, it makes sense to want to ensure they really want the position since interviewing and onboarding are very expensive for a company as well as having someone leave with a bunch of institutional knowledge built up over the years.

It’s also probably the easiest question in the world to totally bullshit with just a vague grasp of what the company does.

“One of the things that really excited me about <company X> is the problems you must face dealing with <such large amounts of data, securing your system, supporting N users on the app at once, syncing data across data zones>” etc etc. or “I think it’d be really neat to work at a company in the <healthcare, finance, whatever> industry because I’ve always wanted to <improve lives, learn about high frequency trading>”

And even if it’s an easy enough thing to BS it should also be genuinely easy to find an exciting reason to want to work somewhere in our field.

A better question might be to make the company attribute hypothetical. Something like: "if you could propose a rule for all employees at the company where you want to work, what would it be." You'd definitely need follow-up questions and promoting, but it could be way more insightful.

My group at work is largely full of people who believe in what we do, enjoy the work we do, and feel an obligation to our users.

For many groups, the only thing more important than getting the right people in, is keeping the wrong people out. Your talented employees can probably quite easily find a lateral move. Hiring the wrong person can lose you several of your most talented team members.

I think you could probably massage that answer though “well standups aren’t my favorite thing really, it’s sort of a lot of context switching and a lot of times the conversation isn’t relevant to me. But I understand it’s generally more for the managers than it is for me.”

And for the latter one

“I like a very professional, cordial, and relaxed relationship with my coworkers. I tend to do my own things on weekends which leaves me refreshed and happy to see everyone on Monday.”

It’s just a bit smoother than “the last thing I’m going to want to do is spend _more_ time with you”.

Instead of emphasizing the drain it’d be to spend more time with your coworkers, emphasize the benefits of coming back refreshed on Monday :-). They’re the same answer but one’s more positive.

“I undid some hacks that Dave who left two years ago had put in the test suite because he just really liked stacking Python decorators, and they were getting too complex to figure out when some test broke. I updated some documentation that hadn't been touched in 5 years and was terrifyingly out of date. And then I sat in for two useless meetings because leadership really likes holding them.” is not likely to communicate any positive messages, but it is the reality of what a lot of us do at our jobs instead of solving complex race conditions in elegantly designed distributed systems.

Not saying that that might not contribute some kind of signal to a potential interview, but it's very likely to be one that labels me as incompetent :)

e.g. for #2 they're not actually doing due diligence on your current employer, they want to know about the sort of environment you thrive in, so talk about what you're looking for in a company's culture.

As you said, #3 is a chance to give lots of real world examples and play to your strengths.

For #4 they don't need to actually know about that specific project, they just want to know how you handle projects, so if there's a way you can answer that from your own experiences then that's fine.

I'd say your #5 is fair enough too. At that point that's probably the sort of thing you'd want to turn around. e.g. ask if they have hired people who have come from self-employed backgrounds, and if so how did they handle it.

If you have hobbies, volunteer work or other experiences that are more recent, I would not hesitate to say something like "Because I've been self-employed for two decades, the last time I had a disagreement with a manager was 1999. But last month at (local non-profit/recent group project in a class I took/whatever), I disagreed with how the people in charge were handling X and..."

They probably really don't care if you talk about work experiences per se. They want to know things like how you handle interpersonal friction.

Can you play well with others at all? Or have you been self-employed for two decades precisely because no one in their right mind would want to put up with you for more than two weeks?

So try to think about some recent experiences that would provide some kind of evidence supporting the idea that, yes, you can get along with other people and they won't regret inviting you into their social group known as X Department at Google.

I will add that they probably frame it as work related because they don't want to hear about your spat with your wife or teenaged son. So don't pull out personal anecdotes of that sort. The social setting needs to be comparable to a work relationship.

Also, some people are bad at guessing expected answers.

I'm definitely not everyone's "cultural fit" (a dog whistle term for legitimizing discrimination imo), but for companies that appreciate what I bring to the table, it's all the better.

Honesty and trust are definitely top values for me, and people who can't handle respectful and frank discussion can't handle me, so better to find out immediately, as early in the process as possible.

If you can’t name a single issue in your org that smells fishy to me, but an honest answer indicates an open culture and decent people.

Only experiencing it will tell you the truth...

For you as the interviewer, an answer like that is a great hook to ask "well, what does matter to you? What would convince you to work here?" It'll tell you a lot about fit or not :)

I want to chill and pace myself here. I'm looking to do 30+ hours of work per week on my projects outside of this dayjob, so it's important for you not bother me too much during my 40 hours here.

My only goal is to build a company on the side. I've turned down better jobs just so I work in an environment and with coworkers that I fucking hate, so that I am extremely motivated to do whatever it takes to have my own business. Overall this job is just a paycheck - a necessary distraction to maintain a reasonable standard of living, and completely irrelevant to any of my goals in life. If contract work didn't cause too much stress/friction for me, I would be doing that for money. I hope you understand that I want to see you fail, this company fail, and any project I'm on to fail miserably. I will do the work you assign me according to spec, but please remember to leave me the fuck alone.

So... when can I start?

No culture questions (I’m happy to manage all kinds of people). No technical challenges. The more you do this, the better you get at it.

I would love to be wrong, but interviewing still seems more magic than science. I've interviewed for positions and think I did great, and did not get a job, and for others, did poorly and did get an offer. On the other side, I've hired promising candidates that turned out terrible, and questionable candidates that turned out great.

There are some things you can do to filter out the worst of folks (e.g., a fiz-buz test or evaluating basic communication skills), but these generally just filter out the bottom 20%. It's maddeningly difficult to differentiate between the middle 50% and the top 20%.

When you occasionally find a rock-star, they generally know they're a rock-star and demand rock-star pay. The top 5% of candidates is rarely 2X more productive than the top 10%, but they can demand much higher salaries because they are easier to identify.

> “What most people mean by culture fit is hiring people they’d like to have a beer with,” says Patty McCord, a human-resources consultant and former chief talent officer at Netflix.

From Google's article:

> However, culture fit evaluations can invite bias into your process if interviewers see it as simply a way to assess a candidate’s likeability, rather than how they align with your company’s core values. Assessing culture fit is about much more than whether or not you’d like to hang out with someone at a company happy hour.

Disclaimer: I work at Google; opinions are my own.

;)