Ask HN: How to interview a full-time designer?
6 comments
Yes, do ask them to design something if that helps you, but make sure you're having him/her design something on the fly that's related to your product, not a real problem you're actually having currently. When interviewing for my summer position at Milk, a part of the application process was re-designing a product that was similar to what Oink could eventually be. The exercise is supposed to be theoretical anyway, so this just furthers the amount of time a designer has to think about the problems and to just get to the solution. This could also be a take-home assignment where they allocate Xhrs to the problem.
More than anything, recognize that the purpose of this exercise isn't to see the designer's abilities (you should be able to see this in their portfolio of previous work), but more about how they approach problems and how they think critically. Understand that no one can ever really outline their design process specifically, but should be able to speak to about the way they solved problems in the past (again, see portfolio) and how they came to solutions.
Talk to them about what they view as "good design" and why they think it. A good designer is ultimately a good communicator and should be able to talk you through why they think it's "good" (in their opinion).
Something that might provide insight is this Behance interview of Ben Barry, Facebook's communication designer: http://the99percent.com/articles/7118/Facebooks-Ben-Barry-On.... It's not completely specific, but it will give you insight into the way most designers (in Silicon Valley, at least) think about their jobs and their work.
Hope that helps!
More than anything, recognize that the purpose of this exercise isn't to see the designer's abilities (you should be able to see this in their portfolio of previous work), but more about how they approach problems and how they think critically. Understand that no one can ever really outline their design process specifically, but should be able to speak to about the way they solved problems in the past (again, see portfolio) and how they came to solutions.
Talk to them about what they view as "good design" and why they think it. A good designer is ultimately a good communicator and should be able to talk you through why they think it's "good" (in their opinion).
Something that might provide insight is this Behance interview of Ben Barry, Facebook's communication designer: http://the99percent.com/articles/7118/Facebooks-Ben-Barry-On.... It's not completely specific, but it will give you insight into the way most designers (in Silicon Valley, at least) think about their jobs and their work.
Hope that helps!
I would stray away from an on-the-spot design task. Design and UI is a very research and strategy oriented task. The result of an hour or day long test may not be an accurate representation of what the designer is capable of.
If you like his work, commission him for a freelance project as the test. Give it milestones and specs like any other project. Based on the results - concerning time management, style, ability to digest the specs, etc, you can have a better idea of the applicant.
If you like his work, commission him for a freelance project as the test. Give it milestones and specs like any other project. Based on the results - concerning time management, style, ability to digest the specs, etc, you can have a better idea of the applicant.
Great feedback !
We'll seriously consider hiring him as a freelance first.
this sorta thing is pretty religious, so i'm just going to note the following points:
1. i've known traditional print designers that have jumped headfirst into html/css/js/etc and many of them have left behind artifacts that still live on the web: blogs they've themed, mini web apps, random code snippets, their own web portfolios, etc. investigate these and see if they mesh with what's on their portfolio or what they claim about themselves.
2. formally trained designers have had lots of experience with critique (and, by proxy, one hopes, critical thinking). Lots of crit is bs, but i tend to favor designers who are thoughtful and can back up their work critically or be honest about their motives.
3. you can assign "design tasks" if you want, but you need to be clear about what you're measuring and how your questions elicits that. If you're having trouble picking a design task, try a classic system design task (let's design usb from scratch...) and then see what that measures and if you can find some analog in your own problem domain.
4. the process is not foolproof and you will likely turn away very qualified people regardless of your metrics.
5. be clear about what you're looking for: do you want somebody that makes stuff look totally sweet, or somebody that solves problems well, or a person that codes monday, designs tuesday, and then specs out a new feature on wednesday... be realistic about your expectations.
6. people interpret the term 'designer' in a bunch of ways. lots of people take this to mean: "i can think about user flow, use visio to make wireframes, etc" and others take this to mean "how can this look awesome." others take this to mean "i thought through where a menu tree went in an enterprise java application" and others take this to mean "i make graphics design, do you need a logo?" and some others are like the wolf from pulp fiction, they just fix underspecified problems and are good brainstormers. none of these are better than others, they all are useful for different reasons and end-results.
but what i'm saying is "you want a designer? why?" be honest about what specifically motivates your desire to hire a full time designer. this will make any interview you run easier to evaluate as a success or failure.
1. i've known traditional print designers that have jumped headfirst into html/css/js/etc and many of them have left behind artifacts that still live on the web: blogs they've themed, mini web apps, random code snippets, their own web portfolios, etc. investigate these and see if they mesh with what's on their portfolio or what they claim about themselves.
2. formally trained designers have had lots of experience with critique (and, by proxy, one hopes, critical thinking). Lots of crit is bs, but i tend to favor designers who are thoughtful and can back up their work critically or be honest about their motives.
3. you can assign "design tasks" if you want, but you need to be clear about what you're measuring and how your questions elicits that. If you're having trouble picking a design task, try a classic system design task (let's design usb from scratch...) and then see what that measures and if you can find some analog in your own problem domain.
4. the process is not foolproof and you will likely turn away very qualified people regardless of your metrics.
5. be clear about what you're looking for: do you want somebody that makes stuff look totally sweet, or somebody that solves problems well, or a person that codes monday, designs tuesday, and then specs out a new feature on wednesday... be realistic about your expectations.
6. people interpret the term 'designer' in a bunch of ways. lots of people take this to mean: "i can think about user flow, use visio to make wireframes, etc" and others take this to mean "how can this look awesome." others take this to mean "i thought through where a menu tree went in an enterprise java application" and others take this to mean "i make graphics design, do you need a logo?" and some others are like the wolf from pulp fiction, they just fix underspecified problems and are good brainstormers. none of these are better than others, they all are useful for different reasons and end-results.
but what i'm saying is "you want a designer? why?" be honest about what specifically motivates your desire to hire a full time designer. this will make any interview you run easier to evaluate as a success or failure.
One thing I do is that I show the person our current UI, and ask them what changes would they propose to make things better and the reasons behind their decisions. This really helps to see how the person approaches problems and what kind of design mentally he/she has.
While I agree that a live test like this might be a good indicator, I don't know if the candidate would get a vibe that you're trying to get free work (A bunch of suggestions to try after you show them the door). Maybe using some other site, like a competitor or something unrelated.
If you do ask someone to mock something up, make sure it's very small (1-2 hours of time max). There's a fine line between an evaluation and spec work, the latter of which will drive away a lot of good candidates.
Ask them what they think about the TechCrunch and Kotaku redesigns. If they actually like them, let them go on for a few more minutes, thank them for their time, escort them out, and throw their application away.
Those redesigns are everything that's wrong with young designers (me included). Looks over usability, creative over predictable is the WRONG way to design. Good design is about usability and usability is about predictability, not over complicating what isn't broken for the sake making it different, not hiding elements and loading up info on rollovers just to give the site a "cold, heartless, minimistic, look at me I'm the next Steve Jobs, look", not segregating information pointlessly.
If they can't tell you what's wrong with the designs in detail, let them go.
Those redesigns are everything that's wrong with young designers (me included). Looks over usability, creative over predictable is the WRONG way to design. Good design is about usability and usability is about predictability, not over complicating what isn't broken for the sake making it different, not hiding elements and loading up info on rollovers just to give the site a "cold, heartless, minimistic, look at me I'm the next Steve Jobs, look", not segregating information pointlessly.
If they can't tell you what's wrong with the designs in detail, let them go.
I agree with the usability issues.
We hired a freelance designer few months ago, all their work was gorgeous but the usability sucked.
We hired a freelance designer few months ago, all their work was gorgeous but the usability sucked.
that's just, like, your opinion, man.
i personally value anyone that can articulate what they think about those redesigns, either like or dislike, thoughtfully without buzzwording me to death and without parroting rhetoric.
you're describing "culture fit", which, while valuable, isn't a good metric for evaluating a designer's skill at... designing.
i personally value anyone that can articulate what they think about those redesigns, either like or dislike, thoughtfully without buzzwording me to death and without parroting rhetoric.
you're describing "culture fit", which, while valuable, isn't a good metric for evaluating a designer's skill at... designing.
I was wondering if i could/should ask him to mock or design something (the same way I ask a developer to resolve a problem).
Do you have any advices or resources ?