4 Steps to creating good interview questions

Joe Caccavale

2

September

2020

|

5

min read

X

We're rebuilding the hiring process.

Want to see more content like this? Subscribe to our fortnightly newsletter.

Give me more

WARNING: this is not a list of generic interview questions.

Interview questions we’re NOT going to suggest you ask:

Tell me about yourself.

What are your weaknesses?

Where would you like to be in your career five years from now?

What do you do in your spare time?

Why? Because traditional interviews aren’t predictive of actual ability and tend to be riddled with bias.

Instead, we’re going to arm you with our behavioural science-approved interview questions. 

Our promise to you: you’ll be able to find the best person for the job and candidates will remember and talk about their interviews.

So, what makes a good interview question?

  • Predictive of actual ability
  • Fair and free from bias
  • Targeted at a specific skill
  • Enjoyable for candidates to answer

Good interview questions are ones that simulate the actual job.

Asking about someone’s education and experience might make you feel like you’re getting an insight into their suitability, but these are just proxies.

Education and experience are, in fact, pretty weak predictors of ability.

The research says that the most predictive form of assessment is work samples.

In short, these are questions that test the skills needed for the job by getting candidates to perform small chunks of the job itself - or tasks that closely mirror them.

Even if just for one interview to see for yourself - try ditching any questions about someone’s background and follow the process below.

Here’s how to turn work samples into interview questions...

1. Know which skills your interview questions are testing

Work samples are less biased than traditional interview questions because instead of probing into someone’s background, they focus solely on the skills needed for the role.

Before heading into an interview, you should have 6-8 skills written down. These can be a mix of technical and soft skills.

These are the skills we used for a recent Senior Full Stack Developer Role.
Based on a candidate’s performance against these skills, you can even create skills maps like the one above.

By putting emphasis on skills, you can get a picture of someone’s skill set and see which parts of the job they would or wouldn’t perform in.

It makes hiring decisions less biased and more objective since you’re comparing skillsets rather than personalities. 

Once you know the skills you’re looking for, you can create interview questions that test for them.

2. Pose interview questions hypothetically 

Remember: good interview questions simulate the role. So, start by asking yourself which parts of the job would test each skill (you can group a few skills together in a single question).

For example: a content marketer like myself requires strategic thinking skills - when I analyse the success of my blog posts and look at the data. So, I was asked this interview question: “How do you measure the success of content that you have put out? What does good look like?”

As a general rule of thumb, if it starts with ‘tell me a time when,” it’s not a good interview question.

It’s unclear to candidates whether they’re meant to just make up a time when something happened or come clean about their lack of experience with that specific situation.

Just because it hasn’t happened to them yet, doesn’t mean they wouldn’t be equipped to deal with it.

Pose these sorts of questions hypothetically instead.

Ask them how they would deal with it instead of how they did deal with it.

Our earlier revelation around experience rings true here also - experience isn’t everything… a candidate who has never done something before could be better at that thing than a candidate who does have the experience.

How do you test this? By getting them to actually perform that task in the interview.  

Good interview questions don’t even need to be questions per se.

They could be a mini-presentation, a prioritisation task, a mock client meeting or a planning task - any part of the role that can be picked out and simulated.

Look at any projects that are coming up or that have recently passed, since good interview questions simulate real-life, why not use a real-life example?

3. Don’t ask about background or hobbies

Much like experience and education, hobbies and interests don’t tell you anything about someone’s ability.

If your interview consists of this sort of chit-chat, then I hate to break it to you, but you’re being biased.

Yes, there is something to be said for personality.

You want someone with a passion for what they do and what your organisation does...

But assessing how much fun a candidate would be at the work drinks isn’t the way forward.

Replace any sort of culture fit questions with mission and value alignment ones.

Passion and value alignment are working characteristics in their own right and so can be tested just like the other skills you outlined earlier.

4. Good interview questions should have scoring criteria

By using more objective, skill-based interview questions, you make it easier to actually review candidates.

Whist most interview decisions are guided by ‘gut feeling,’ you can now base your decisions on data, rather than bias that doesn’t actually find the best people.

For each of your interview questions, write down a 1-5 star criteria for scoring.

It doesn’t need to be overly detailed, check out the example below of an interview question we asked sales candidates and the corresponding criteria.


At the end of the process, you can add up the scores and there should be a clear best candidate.

If you’re asking good interview questions, giving objective, personalised feedback becomes easy. You can simply tell candidates how they scored against the various skills. 

We actually made a template for giving this sort of feedback.

Good interview questions made easy (with examples)

Grab our free interview playbook. We go through our question creation process in detail as well as the unbiased, more predictive structured interview process we use here at Applied.

We even threw in a ton of real interview questions we’ve used to hire the team.



We pride ourselves on using the fairest and most accurate assessment methods behavioural science has to offer. The Applied platform was made to turn the hiring process into a science so that you get the best person for the job, and have the data to prove it. Browse our resources to find out more or start a free trial of the platform.