I broke the rules and read the article before commenting.<p>The article concludes<p>> The bottom line: it is far less damaging to score 1 rating point lower on Communicate, and it is much more beneficial to score 1 rating point higher on Code and/or on Solve. Talk is indeed cheap, and indeed, coding and problem solving is what you need to show in a coding interview.<p>which is the opposite of what the two current HN responses predict.
I cannot overstate how important communication <i>actually</i> is. A huge bulk of what engineers do is communicate. Design docs, white boarding, tech talks, gathering requirements, challenging leadership, organizing a team, code reviews, peer feedback... Nearly everything engineers do requires communications skills.<p>Being able to connect with and clearly communicate to your interviewer will <i>absolutely</i> increase your chances.
This is an awful article. I would rate someone as having poor “solving” skills if they demonstrated this level of thinking.<p>Most fundamentally, of course the data shows that “would hire” correlates to “appears to have the skill this interview is specifically intended to assess”. These are not independent variables, nor are they objective measures. Don’t waste your time analysing them as if they are.<p>From a candidates point of view, the question is “what can I spend time on to improve my chances of getting the job I want”.<p>I’ve never met anyone who’s preparation is analogous to the scenario presented, “I decided to convert a communications point into a coding point.” Developing complimentary skills is rarely a zero sum game.<p>The data appears to show that unless you are top ranked on two different technical criteria, poor communication skills will prevent you from getting the job. Data-wise, we’re starting to see the impact of having a variable that isn’t almost perfectly correlated with the outcome. That’s less likely to mean “talk is cheap” than “there’s an easy opportunity to improve your performance; you’d be a fool to dismiss it”.<p>I would also suggest that earning a high score for problem solving is going to benefit significantly from good communications skills. Asking good questions and explaining your process, whether or not you could code the solution, goes a long way.<p>So sure, if you are extremely skilled on the technical side, and the company hires based solely on the technical interview, and that company is offering you the best of all the potential roles you could take, better communication skills aren’t necessarily valuable to you.<p>For everyone else: career tip, of course you should be applying for jobs where your skills are relevant to the role, but also remember that general communication skills are valuable.
"Talk is cheap, show me the code."
Any engineer worth their salt will "naturally" communicate by asking questions, appear friendly, or even admit when they're stuck or don't know something.<p>The problem is the reverse is probably as common, but, honestly, equally valuable: if you have a person who simply shows amazing tech skills, and talks when prompted to, I mean, will you reject this person in favour of a less stronger candidate just because that weaker candidate was more outgoing?<p>I mean, in technical interviews you have to be technical and solve your stuff, that's it.<p>FAANG system design interviews actually somehow require some clarifying questions to clear out ambiguous requirements or probe for missing data, but, I have seen and witnessed people immediately tabling X, Y and Z assumptions and go "assuming these....." and immediately craft a perfect solution - pass with flying colors.<p>The rest will be gauged in behavioral interviews and in the final decision meeting to put forward an offer or not.<p>But in technical interviews? Just solve the thing.
It'd be nice to elaborate what communication in this specific setting entails.<p>From my experience:<p>- being willing & able to "bring along" the interviewers, thinking out loud<p>- being easy to follow (clear phrasing, logically chained etc.)<p>- asking for help/input when needed<p>Without defining what exactly 'communication' means it's impossible to measure. Even then, it's highly subjective - that doesn't make it less important though.
There is nothing scientific about the setup, it implies bias front and center, the conclusion is completly flawed and the title doesn't even state the result correctly:<p>Indeed, at best, this is not about how well a candidate does at interview depending on how good at communicating they are.<p>This is about how companies rate canditates on communication depending on if pass their interview or not.<p>You can't post a comment on HN without people obsessivelly asking proof about your opinion, numbers, studies, data...<p>And this reach the frontpage. How ironic.
If I'm understanding this correctly, this is based largely on the results of <i>mock</i> interviews. The author says that interviewing.io "has hosted and collected feedback from over 100K technical interviews, split between mock interviews and real ones" but does not say how many (or if any) of the interviews these conclusions were based on were from mock interviews. My understanding is that interviewing.io is largely for mock interviewing, although some interviewers may funnel you in to their real hiring pipeline if they like your mock interview.<p>I think mock interviews are great, but do we have any solid data on how well the "advance to next round" flag for a mock interview correlates to advancing to the next round in a real interview?<p>If I'm conducting a real interview, I'm going to be thinking about what it would be like to work with this candidate on a daily basis (or at least have them somewhere in my organization). I would think I would be more concerned about communication skills in a real interview, even if it's unconscious. With a mock interview, I'm in no danger of working with this person. I can be focused completely on their technical solution without thinking at all about what it would be like to work with this person day to day.
Many posting responses here are missing that this topic is constrained to _technical interviews_. Not "does communication matter in general" nor even "does communication matter for a developer to be effective".
I wonder how terrible the 4-code 4-solve 2-communicate people <i>actually</i> were at communicating. They were able to communicate their code and solutions to the interviewer after all. There's a great chance that the people who really <i>are</i> poor communicators were rated lower at least on the solve scale.<p>Looks like as usual, <i>"more research is needed"</i>.
So what does communication mean in this context? And is that a shared meaning among all participants?<p>Unless it's clearly defined I don't see how anyone can measure it in a meaningful way.<p>This seems to be treating communication as some kind of qualitative abstract.<p>"On a scale of 1 to 4 stars, how much did your spouse love you last Wednesday?" That tells me more about the esteem of the respondent than it does about the state of the person we're trying to evaluate.
I think the article is illuminating. When I was actively interviewing, I found that I struggled in code interviews precisely because I was trying to be engaging, but it interfered with how I was actually solving the problem.<p>I think I would have done better if I had taken the advice to communicate less during the code review and explain after I solve.
This data might explain why technical documentation is so poor or non-existent. I give bonus points to good communicators, because good communicators are so rare.
so long as we are passing requirements from one person to another, communication will remain a priority skill for coding. The only circumstances where I can see communication becoming a secondary skill is a) solo coder or b) repeat process / build
> Does communication matter in technical interviews?<p>Who thinks up such a question? It’s like asking, “Is the sun hot” and then writing a blog post investigating the obvious.