How we approached (video) interviews
The truth is that interviews oftentimes is a stressful and displeasing experience. Now insert the video parameter there, too.
One of the best commercial digital products I’ve always admired is Google Maps. Spoiler: it’s not because I am Google fanboy. What I like most about that app, is the simple fact that whenever I use it, I am always under some kind of pressure; either I am driving at unknown places during vacation or I am abroad exploring and touring totally different countries. Whenever I use it, I want no mistakes. I just want reliability and precision.
Since I am fortunate enough to witness how these kind of products are made, how the teams are working on them and even work with some people that built them, I could not help myself, I wanted to find a similar problem to solve: design an interface for human beings under stress, with a goal to remove friction and reduce stress.
Stress mode: on
One a bright morning at a Workable Summit, we were told that a new team will be formed with one goal: build a new product to allow candidates perform asynchronous video interviews. The business case is solid; on the recruiters side, saving time so they don’t have to phone screen candidates. On the candidates side, provide a reliable platform to record your answers and send them to the company they applied to.
We wanted to examine how the candidates will record their answers first. We started working on this with a design sprint. We talked with several real candidates, even some that had performed a video interview before. Fun fact, you can check tons of feedback about the whole video interviews concept publicly, for instance at this subreddit.
In short, the outcome was pretty simple. The number one problem we would have to solve, as far as the candidate side is concerned, is the stress. I know, it sounds kinda generic and vague, but picture this: You are a candidate and you have to record and submit a set of 5 questions via a browser. You might be afraid of a trillion things such “this is my first time doing this how this platform even works”, “what if my internet fail me”, “how will I look normal talking to a computer” — the list might go on for a few days 😄
Design comes into play
Since that list is pretty enriched, it was obvious that we had the problem outlined. That’s the right moment to jump into Sketch and solve that problem, right? Wrong.
It was too early. We gathered more feedback from past candidates who did video interviews in the past and grouped it into prioritised concerns. That feedback gave birth to a principles set, that would guide us when we built the application.
Note: by ‘us’, I mean the engineering team too. I am a fan of being extrovert designer; I want to document rules, rationale and my general thinking. I think it makes it easier for everyone to keep up and be on the same page.
So what about that principle set? It was a light, pleasant, three points set.
1. Make feedback always be visible
The first and toughest point. We needed to allocate a reserved real-estate on the screen to provide platform feedback to the candidate. This feedback, is not only about error messages, it’s about guiding the user throughout the whole process. It’s more about providing a north star.
Of course it plays well with error messages too. Assuming the person will get the concept, that a message is always there showing me what’s going on, if anything comes up (i.e technical issue?), this will be the spot they’ll look for feedback.
2. Be picky about the options
During high stress levels, what you choose to expose to the candidate is extremely important. Remember the Google Maps example at the begging of this story? Google Maps is doing an excellent job structuring the information in such a way, that you’ll have your route in the spotlight throughout your experience.
The image below is a great example on how we achieved this for Video Interviews. This is the scenario when the candidate has the max number of CTAs displayed in the interface.
3. Seperate screen into logical groups (and group relevant actions)
Video interviews in general, is a fairly new concept. That means, there is no known path to follow, and make the interface understandable by commonality. What I chose to embrace, is to seperate the interface into logical groups — and obviously to group them together. By doing this, we would reduce the learning curve of using the interface.
We ended up having Navigation, Questions & Interview controls. So each one of these three groups represent a specific pool that similar actions should be close coupled.
Outcome
Stress is something difficult to measure. Business metrics, retention or adoption is much easier. The stress factor is difficult to solve partly because it’s difficult to measure and to know what’s producing it. The number one goal, was to reduce the stress candidates may have, by providing an interface easy to grasp and understand. Interviews are by definition stressfull for most people, so there’s only so much we can do.
Moreover, we wanted to make the human behind the screen feel that they are in control of the interface. Recording will start only if they want, the interview can wait for them to be ready, their answers will be sent only if they want them to be sent. We didn’t want to craft a ‘smart’ experience that make choices for you.
Having said all these we consider this project a success since the completion rate of given video interviews is much higher than the industry standard. The industry standard is that ⅓ of the candidates complete their video interview. We managed to rise that number up to ⅔ for our Video Interviews platform.
As a closing thought, we tried to craft an interface as simple as possible to allow candidates give their best self for their interviews. Talking to a machine might feel a bit awkward. Obviously, we didn’t want candidates to feel that; we wanted candidates to feel comfortable and in control. Here’s a snippet of the final shipped product:
