A lot of tech people approach hiring the same way they approach shopping for a new computer. You ponder your needs, you come up with a spec sheet, and you start looking around for that one sweet deal which has all the features you need at the price you can afford. "All we need," goes the hopeful thought, "is to really be clear about the role and responsibilities."
Unfortunately, that's not how hiring or team-building works.
Here's a better analogy. Hiring is, in a lot of ways, a lot like shopping for a new place to live. That, too, begins with requirements (often, let us say, an "optimistic" list). It starts off with the dream scenario:
"we need a place near the #2 bus line, which takes dogs, has good internet, has a fair-trade coffee shop and a vegan bakery within walking distance, has AC, with a roof garden and a view of the lake."
After few weeks or months of searching that list of "must-haves" has usually shrunk, often quite a bit. The realities of what's on offer rarely live up to our hopes and dreams: it turns out there aren't a lot of vegans riding the #2, so some hard choices need to be made. On the other hand the search process itself can reveal a few desirables that weren't on the initial list: "The bike storage in that place is really cool, I think we should add 'good bike locker' to our list!"
At the end of the process you'll end up in a place which (hopefully) is recognizably similar to what you set out to find, but only rarely will it be the place you imagined when you started. The process of looking teaches you more about the distinction between what's vital ("we have to find a place for Fluffy!") and what's not really important ("ah, we'll just by a room AC for the bedroom"). Often, the critical selling points in the final decision emerged during the search itself: it's not uncommon to wind up in a neighborhood that you didn't expect would suit you, or to realize that your initial list of wants and needs was missing a few important pieces.
All of the above is remarkably similar to what happens when you set out to hire someone. It's a journey, often a pretty surprising one.
If you think about this analogy you'll notice one thing which it does not evoke: house-hunting bears little resemblance to things like college admissions or standardized tests. Hiring is about figuring out how different, unique individuals might fit into a team or a project. Skills and prior knowledge are part of that picture, to be sure -- but often not the most important part in the final decision, even though those are the easier things to think about in the abstract.
Just hiring a person is not the same thing as picking a new video card -- while specs certainly do matter, they are far from the representing the complete story. This is particularly true for tech artists and our fluid job responsibilities: even between two different teams in the same company the expectations that come with a given title can vary tremendously. Social skills, design intuition, traditional art chops, and leadership can all matter as much as mastery of Python or 3d math.
When you look at an existing team, what stories jump out at you? It's almost certainly not "X really knows when to use a list comprehension" or "Y's linear algebra is completely appropriate to their career grade." It's more likely to sound like "X is a great at building relationships with the artists, but doesn't know when to say 'no' to low-value requests" or "Y writes very solid code, but is hard to redirect when the plan changes." A good team is not built on LeetCode rankings or the numbers of certifications people have acquired; it's a mix of skills and perspectives and growth opportunities. If you try to reduce all hiring decisions to some kind of skills matrix you're ignoring the most important things you already know about what makes your team work now.
This is not to say there's no value to understanding your candidates' technical skills and experience -- only to emphasize that figuring out the technical bar is generally the easiest and least interesting part of finding the right people to grow to your team.
One way in which a candidate search really resembles a housing search is the way both of them stress your ability to manage choice. At the beginning of a housing search you're applying a lot of filters -- you don't have the time or energy to visit hundreds of places. But the more filters you apply, the smaller the pool of possible places becomes. Nowadays you can preview this in real time on any house-hunting website: the transition from overwhelming abundance to a worrisome lack of options happens really, really fast.
The same thing applies to the hiring process. That dream candidate -- the one who has done exactly this role before in precisely they way you want it done and who is excited to do it again without a huge pay bump -- probably does not exist. The future colleague you're going to end up hiring, on the other hand, has all sorts of qualities you didn't even know were important until you saw them in action. A good interview process keeps you open to the qualities you uncover along the way, keeping the door open to the values and opportunities that come from a broader range of candidates.
So, when in doubt you should err on the side of a smaller must-have list. Over-specified jobs tend to turn off candidates from underrepresented backgrounds, which is reason enough to keep your list of requirements short. Moreover a glut of bullet points also translates into a more homogenous skill sets and professional backgrounds, hampering the interview team's freedom to re-envision the role as they meet more people.
A light touch with filters also gives you a chance to manage your risk. The quest for the unicorn candidate often turns suddenly into a panic hire driven by the need to fill an empty chair or the fear of losing headcount. The pickier you are in the filtering stage the longer it will take you to get a solid sense of what the market is like and how realistic your expectations are.
Crafting a good job description, like finding a good house, requires a reasoned mix of specificity and openness. Too broad a description and you'll end up drowning in low-yield resumes -- but add too many bullet points and you'll scare off plenty of people who could have been great additions to your team.
All that said, a more broadly specified job posting will absolutely create more work for the hiring team. However, that's a feature, not a bug. The house you didn't know you wanted often turns out to be better than the one you imagined. Likewise that ideal candidate in your head on day one is just an uneducated guess about the right way to fill the role; once you've had a chance to learn from some real flesh-and-blood candidates you'll have a much richer set of tools for imagining how that role could and should be played in practice.
In a house hunt, there comes a time when the binary process of filtering out places you don't want transforms into the nuanced process of choosing between options with unique, and not directly comparable qualities. This part of the process calls for more reflection: is there really some objective way to balance, say, that vegan bakery with that really serene lake view? The final round is a time when you learn a lot -- often unexpectedly -- about your own priorities and values. Every hiring decision turns on individuals, with all of their unique (or even idiosyncratic) individual strengths and weaknesses. No matter what you think you're hiring for, there is no way to remove the human element of this decision.
It may sound admirably meritocratic to give every candidate some kind of objective test and to pick the highest scoring candidate: but do you really want a technically proficient jerk? Or a team full of people who excel at logic puzzles but can't deal with the chaos or productions? Conversely, you might hire based on how well a candidate jibes with the interview team. Both figuratively and literally, somebody who speaks your language looks superficially like an easier hire. But tech skill and language fluency don't correlate, good interview manners don't always equate to good teamwork, and "culture fit" can really mean "groupthink". Even your desire to diversify your team's demographics doesn't guarantee that you'll actually get the kind of candidates you want. Above all, a good hire will literally never reduce to only one of these dimensions: your team will not prosper without technical skills, social skills and diverse perspectives.
To make things worse, you will be forced to try to asses your candidates in ways that are ultimately pretty flawed. Tests and interviews can be gamed: the internet abounds with advice on how to ace a Google interview or how to maximize your LeetCode potential. But if you read them they often boil down to either (a) an exhortation to just "become awesome" or (b) a collection of parlor tricks that any good interviewer has seen a million times. There's a place for these things: when you're first starting out in the field it's good to know how to carry yourself in an interview. However when the advice morphs from here's how not to shoot yourself in the foot to here's how to fool your interviewer, things get dicey. Let's face it, nobody has fallen for "my biggest weakness is that I work too hard" anymore. Interviews are a weird mix of awkward social event, improv class, and Spanish Inquisition. And -- alas! -- they are the best tool we have when trying to grow our teams.
This will never not be hard. But you can make it easier by according the interview process with the attention it deserves. In our business it is way too easy to view hiring as a distraction from the "real" work of delivering tools and supporting teams. Far too many places treat interviewing as an afterthought, when in fact it's the foundation of a good team. Good hiring practice produces a good team -- being too busy to hire thoughtfully, on the other hand, guarantees a future in which you are constantly firefighting and too busy to hire thoughtfully.
The rest of this piece centers on concrete suggestions about how to do better at interviewing. Before diving in to the details of what happens in those few intense hours, though, take a moment to think about the best example you have of the outcome: your own experience.
We all know that actually living in a new place teaches you how many things you missed during the house-hunt: like "that #2 bus is always late and crowded" or "wow,, those outlets in the office are really poorly placed" or, on the plus side "I've lost 10 pounds now that I bike to work instead of waiting for the #2!" As you consider new hires you should try to be a perceptive observer of current and past team dynamics: what mixes of skills, personalities and backgrounds have worked best ? What seem to be best indicators of an effective group or, conversely, what are warning signals? Hindsight is a wonderful teacher. There's no better to plan for the future of your team than to be well-versed in its past.
And now: on to the mechanics