Why we don’t give take-home assignments when interviewing designers, and why you probably shouldn’t either

It’s a controversial topic, but that controversy escapes many hiring managers: giving homework to design candidates. Many are doing it, and some companies are famous (or notorious) for their tests’ depth. At AppDynamics, we’ve hired over 20 designers who have turned out fantastic — and we didn’t ask a single one to do work outside of the scheduled interview time (we did do that for one round of interns and dropped it again).

I’ll start by acknowledging the reasons why someone may want to do that. You want to assess someone’s design and collaboration skills on a topic you’re intimately familiar with — preferably your very own product. That’s a lot easier than trying to do that by assessing someone’s portfolio because you don’t have to get familiar with the problem space — you already are the expert. Or maybe there were holes in the portfolio for critical skills you’re looking for and want to dig deeper. Perhaps you have doubts about how much a candidate contributed to a project on their portfolio. Or the candidate is coming straight out of school and doesn’t have a meaningful portfolio. Some of these reasons are valid, but the assignments need to be tailored and narrow in scope. I’ll get to that later.

Or maybe you’re a huge corporation recognized as a design Mecca and need a way to churn through the enormous influx of applications. A standardized design exercise lets you quickly compare candidates. In that case, you can get away with anything you want, but it very much shifts the burden to the applicant.

It’s not fair

Let’s talk about why asking for extra take-home work is almost always a bad idea. First, there is the notion of fairness. A job interview is supposed to be a give and take: you need talent, and the candidate wants a job. It’s supposed to benefit both sides, and you should treat it as such. Asking a designer to take on a whole new assignment they’re not getting paid for isn’t fair because any designer worth their salt will strive to do a great job on a design project. Nobody wants to half-ass it, especially with a job offer on the line. If you know anything about design, you know that means these things don’t only take skill, but they take time — time that some candidates may have more of than others. At this point, your selection pool skews towards candidates with spare time or who are desperate for a job, not necessarily the most skilled ones. It also favors the privileged because the underprivileged can least afford to spend time on something where there’s no guarantee of a return. It’s a subtle form of discrimination, and if you’re talking about wanting to hire diversely, this is something to consider.

The last point I want to make about fairness is that it puts designers in a bad position: they either do the work or justify why they won’t. The Valley is small, and the last thing a designer needs is for an influential person to conclude they’re either lazy or cocky and share that opinion with their network.

It’s unnecessary or, worse, backfires

If fairness doesn’t compel you, the second reason should: it’s mostly unnecessary and counter-productive. To strive as a business, you need to attract the best talent, not repel it. Designers generally hate doing extra assignments, especially if they’re applying for multiple jobs and each requires them to do one.

Experienced designers already have a portfolio of their work that you can judge them on. A good portfolio project explains the problem, how the designer approached it, shows artifacts produced along the way, and what the result looked like. What are you trying to assess that’s not already there? The reason I hear from hiring managers asking for take-home design work is that it gives them an “insight into the thought process” or “how the candidate breaks down a new problem.” That’s just lazy because you can get the same insight from a portfolio if you familiarize yourself with the project’s problem space instead of demanding the candidate does that for your domain. I would argue it gives you an even better insight: if a portfolio project makes you understand a solution from an unfamiliar problem space, that already tells you that the designer is good at storytelling.

Ultimately, your hiring process may turn away the most experienced and brightest designers. Most of the time, you won’t even notice because once word gets out how the hiring process works at your company (and people will be asking about this on Blind and Glassdoor), these individuals won’t even apply. Finally, if you’re not sure what exactly the candidate’s contribution to a project was, ask for references and check — it’s easy and effective.

If you’re gonna do it, do it right

Let’s consider the cases when take-home assignments may be beneficial and how to do them right. Let’s say you’re looking for a designer with a specific skill set, and that skill isn’t represented well in their portfolio. Maybe you need someone brilliant at creating vector graphics, and nothing in their portfolio hints at them being able to do that. It’s OK to ask that person to show you an example that shows they have the skill you need, and if they can’t come up with one to create something new. As long as you keep it narrowly defined to address the gap, this is perfectly fine. Don’t fall into the trap of having a predefined assignment that you assign to everybody, no matter what questions you actually have. Create a new assignment for each candidate where you have doubts and only test that skill.

Don’t do broad end-to-end scenarios. It’s very tempting to create an assignment that tests a complete flow. That is very time-consuming, even if the expected outcome is “only” a high-fidelity mock and a write-up. A good designer will be thorough when asked for a high-fidelity mock. They need to understand the problem first, research existing solutions, and verify their assumptions with someone before even opening a design tool.

Avoid ethical dilemmas

For junior positions where candidates won’t have a very relevant portfolio yet, the candidate may be happy to do a project, but don’t just hand them an actual design problem that you’re working on in your company. I get it, it’s fresh on your mind, and you understand it very well, but that quickly gets into ethically questionable territory. What if you like something about the proposed solution but don’t hire the candidate? Would you feel comfortable using it without giving credit or paying for it? How would the candidate feel about that? It basically amounts to asking for free work, even if you don’t plan on using it.

If you must make them do a project, come up with one related to your space in terms of complexity, but is clearly more generalized. Or why not go a step further and tailor the assignment to the candidate and give them a problem in a space you know they are familiar with or interested in? Chances are they will be more enthusiastic, take much less time, provide better results, and they can even use the work in their portfolio.

I hear your objection: “But how do we find out how quickly someone can adapt to our very unique space and start producing great work?” The truth is you’re probably not as unique as you think you are, and designers with the right mindset don’t have problems adjusting to a new space. Look at their portfolio. Is it varied? If they’ve delivered results for different industries, in different mediums, at different stages in the product cycle, and with different tools, they already demonstrated that they’re capable of adjusting. At AppDynamics, we’re dealing with a very complex product in a highly specialized domain. Besides one designer who came from a competitor, nobody had directly relevant experience in the APM/DevOps space. Guess what, it didn’t matter. Everybody can learn about the product, and a lot of previous knowledge is transferable in some way or another. Bringing in varied experiences is extremely valuable as it safeguards against tunnel vision. If you need someone who can hit the ground running from Day One, you waited too long to start the hiring process.

Be upfront and transparent

One more piece of advice: be upfront about the hiring process, either in the job description or when you first talk to the candidate. If a take-home assignment is required, state that at the very beginning, so nobody has to waste time going through multiple rounds of interviews if the candidate isn’t interested in doing it. If an assignment isn’t required, but you want to test a gap, make that clear. Set the candidate up for success by being transparent with what you judge them on. Don’t make them guess — be very explicit in what exactly you hope to learn and what you would like to see.

To summarize, take-home assignments are generally a bad idea: they’re unfair, mostly unnecessary, and can prevent you from attracting good talent. If you must do them, be upfront about that, don’t use something you’re actually working on, and be as narrow in scope as you can.

If you’ve weighed all the pros and cons and still come to the conclusion that only a take-home assignment lets you genuinely assess a candidate, then maybe you should go all-in on that notion, and instead of hiring for a full-time position, make it a contract-to-hire job. At least that way, the candidate is getting paid for the work, and it gives you way more time to assess if a candidate is a good fit, but you already know that this would rule out the experienced and most talented designers because they have options. What makes you think a take-home assignment would be any different?

Image credits: JESHOOTS.COM on Unsplash, Skitterphoto from Pexels, Skitterphoto from Pexels, Trinity Kubassek from Pexels, Marc Schulte on Unsplash