How to answer
Open-Ended Interview Questions

As part of my tech resume writing services, I spend a lot of time on answering questions about interviewing from clients who landed interviews.

One that comes up a lot is how to interview well, and more specifically how to answer open-ended interview questions (often asked at FAANG, etc…). There’s mostly generic/vague advice online, which you probably found hard to apply.

So I wrote this step-by-step guide with everything you need to know on the topic. This is an article you can keep referring to, so you can get better at this skill which will serve you for your entire career.

This method is based on my 14 years recruiting experience, especially for Google, where part of my job was to analyze interview performance.

The purpose of an open-ended question is not to get a final answer. It is to get a thought-process.

You're forced to expose your actual chain of thoughts, because:

• You can't use prior knowledge only.
• You can't predict which question will be asked.

The experience can be nerve-wracking, especially if you're new to it. You're already in a stressful situation (interviewing), and you're essentially asked to improvise. You have to think about the answer and communicate it at the same time, which is a lot for your brain to process so it removes all posturing. You're exposed and you have no other choice than to reason out loud.

These interview questions are much more telling than “Where do you see yourself in 5 years?” 😉

Among the many types of interview questions you'll encounter, the simplest split is open-ended questions vs. closed-ended questions.

Closed-ended questions have a finite set of valid answers: yes/no, a date, a number, a name. "Have you used Kubernetes in production?" is a closed-ended question. They're efficient for screening factual job interview questions, but they don't reveal much about how you think.

Open-ended questions can't be answered in one word. They force you to reason out loud, structure a position, and walk through trade-offs. That's why they dominate the harder job interview questions at FAANG and other competitive employers: among all types of interview questions, they're the most reliable way to evaluate how a candidate actually thinks.

Behavioral questions

Behavioral questions are based on past experiences. Their goal is to trigger the memory of an event, which you will then articulate as an answer. By walking the interviewer through what happened, you expose your behavior in a certain context.

They often start with Tell me about a time when… Here’s one:

That's a classic question used at FAANG to evaluate someone's ability to do what's best for the company despite the hierarchy. This fits Amazon's "Have Backbone" principle, or Google's "Do The Right Thing" rubric.

Situational questions

Situational questions are made-up scenarios. They’re in my experience the hardest to answer, because the context could be totally unfamiliar to you. It’s like a simulation lab.

Here’s the hypothetical twist on our previous question, so that you can see the difference:

The context is much narrower, so that situation likely hasn't happened to you yet.

Typically, competitive companies (FAANG, etc…) will use both types of questions during the same interview, so that they can confirm that past behaviors are consistent with potential (future) behaviors.

Even though these 2 types of questions may feel different, the methodology and assessment are the same.

Understand that big companies take their tough open-ended interview questions extremely seriously. FAANG have detailed interview questions repositories with vetted questions to select from. Questions sometimes require approval from the recruiter so that interview feedback is deemed valid. I am not joking: I have had cases where new interviews had to be redone because the wrong set of questions was selected. (nope, candidates weren't happy...)

It's also not just one question: they also have pre-calibrated follow-up questions. They do this to stress-test your plan and to allow you to elaborate. You will give your first "main" answer, and the interviewer will guide you through digging deeper.

You shouldn't see these follow-up questions as a challenge, unless you're forcing the interviewer to ask basic questions on obvious details. They're mostly here to help. My advice is to think of these interviews as a conversation, rather as a Q&A.

Smaller organizations may be less sophisticated, but my advice is to prepare as you would for a top player.

This section is super important, because once you understand how you're evaluated, it will make everything else much clearer.

I can't go over the specific evaluation rubric of each company, and I don't need to. They all tend to gravitate around 3 components: Structure, Complexity and Logic.

Let's go over each of them in detail, so you can visualize the interviewer's checklist.

(1) Structure

The most obvious part is structure, which comes down to these 4 questions

• Situation: Do you understand the problem and its root cause?
• Task: What is the most impactful set of solutions to the problem?
• Action: How do you implement your plan, and with which resources?
• Result: What results do you expect and how will you measure them?

You've probably heard of the STAR method, which helps you organize your answer in successive steps.. That part is actually well covered online, so I won't elaborate on it with this article. No myth busting here: it does work. Unfortunately though, most of the general advice stops here. But we'll go much deeper 😉

(2) Complexity

During my previous career as a recruiter, I had to analyze and document interview feedback to support candidates in front of hiring committees. What I found is that most candidates understood the structure part, and most of the difference was made by their ability to handle complexity.

During interviews, complexity is 3 components:

• Timeframe or your ability to integrate short, mid and long-term scenarios. Are you able to come up with different solutions/actions/results for all these timelines?
• Scale, which is your ability to navigate between the bird's-eye view and the granular details. Can you talk about the general principles / trends, and then suddenly switch to the specifics of the implementation with precision?
• Contingencies: Can you anticipate issues and have contingency plans for them? This is your ability to think like a chess player and plan several future moves based on different outcomes, and course-correct when needed.

Each of these bring an additional dimension to the basic structure, which creates more depth to your answer. This is the part top candidates nail.

(3) Reasoning

The last part is the quality of your reasoning itself. There are 2 components to it, which I'll call critical thinking and support.

Critical thinking is your argumentation. It's whether you made reasonable claims or statements based on the given context. Are you solving the right problems and making relevant hypotheses?

Support is your ability to back your hypotheses. It is a bit tricky: on your resume, you're told to quantify achievements with metrics, but with open-ended questions you won't have those.

Interviewers need to see that you can deduce/infer useful data, or use logical statements to confirm your claims. In the more abstract cases (like the Q&A example at the end of this article), you should at least explain why an action should lead you to the desired outcome.

For more concrete cases (especially for situational questions), you can support your argumentation with past examples, which is what my “Story bank” technique (see below) is perfect for.

Now that you know what you're up against, let's begin your training, young Padawan! 🏋️‍♂️ how do you actually prepare for these, if you don’t know which question will be asked?

I'll give you the method that got me my job at Google, which I kept recommending to candidates.

The key is to train a thought process, instead of specific answers. The best way to get better at this is to focus on each aspect (structure, complexity, and reasoning) individuallyfirst, before putting it all together.

I'll give you an exercise, with 3 levels you can clear.

Level 1: Structure (STAR)

For the structure, you need a framework that will help you organize the steps of your problem solving.

The most well-known is STAR (Situation > Task > Action > Result), but there are others (PAR, CAR, SOAR, etc…). They’re all essentially the same thing: pick one. What matters is that you can visualize the steps.

As a quick reminder, here's how you should think of it:

Here's the Exercise:

This will eventually help you internalize the structure and instinctively think in terms of steps. This is essential to free up cognitive load, so that you can focus your brain power on complexity and reasoning, rather than structure. I did not invent it: this is how musicians prepare for improvisation. They "hard code" scales, patterns and musical phrases, which then come out naturally so that they can focus on creativity on stage.

Level 2: Complexity (Timeframe, Scale, Contingencies)

Training for complexity is a bit harder, because it's less linear. I made a simple diagram to show you where each of the components would fit within the STAR structure (see below).

To train on this, use the the same exercise as before, but this time, add more details on one of the 3 components of complexity (Time, Scale, Contingencies).

Work on timeframe first: you can think of that as adding several Action > Result loops (for short term, then mid-term, then long-term) instead of the unique one you had within the original STAR structure.

Then focus on scale, which fits best within the "Task" step. You can split it into (i) the overall strategy and then divide it into (ii) 2/3 areas of implementation. There's no limit to how complex you can go with creating more "sub-parts", but start by using the simplest version possible first, and complexify later.

Once you're comfortable with timeframe and scale, add the contingency part after the Result step of your STAR system. You should ask yourself "What could go wrong?" and answer the 1/2 first issues that come to mind with (a) problem statement, (b) solution, and (c) expected result.

Level 3: Reasoning (Critical Thinking, Support)

It's hard to train your reasoning with a specific technique. Doing the exercise itself trains that muscle and you should find logical connections more quickly over time.

Here's are the 3 ways to handle the support part:

1. Every time you can think of a metric that could serve as a clear proof, either (a) deduce it logically(with "napkin math") or (b) make an assumption. Explain your reasoning, but don't dwell on the number: what matters is how you use it within your thought process.
2. You'll find that in many cases (especially for "Leadership" oriented questions), justifications are more of a "gut feeling". That is fine: that gut feeling comes from experience and your brain's analysis of past situations. If that's the case, outline the expected cause and effect of an action. (This is what I've done in the Q&A example below, which doesn't include any tangible metric).
3. For situational questions, there is a standard and expected way to use evidence. Evidence is the key component of a good answer for these, which is why I insisted on defining the types of open-ended questions above. I'll give you my secret weapon: the Story Bank (see below).

Ask follow-up questions

If you follow the STAR method, you know that assessing the situation (the problem at hand) is the first step, and follow up questions are a great tool to gather information.

2 small tips:

• Make sure to ask questions that uncover more information. Don't ask questions for the sake of it.
• If you can't think of a useful question, simply rephrase/reframe the problem as a question to get the interviewer’s approval. This will confirm that you’re on the right track while still complying with the "request for information” step.

Make Assumptions

You might need to use metrics, volumes, scales, proportions, etc in your answer, for which you don't know real world numbers. If that's the case, make assumptions and tell the interviewer that "your reasoning takes X as a base to measure Y". >Again, they evaluate you on your reasoning, so the actual number doesn't matter.

Take your time to answer

The third advice is to take your time to answer. Interviewers do not expect you to answer within seconds, but when it does come they do expect your thoughts to be organized. It also helps with perception: someone who pauses before answering appears more thoughtful than someone who rushes to answer.

Think out loud

Don’t try to build the whole answer in your mind before answering. Instead…

(1) Create a rough plan in your head. (2) Then walk the interviewer through your reasoning while adding complexity. Yes, it's a hard gymnastic to handle without experience (hence my training recommendation above), but once it becomes natural it reduces cognitive effort. You get the best of both worlds: a well-thought out structure (prepared mentally), and complexity + quality of reasoning (thought “out loud”).

I tried to write the answer part above in one go, so that it can feel more realistic and less "polished" than a carefully written answer. If yours is within that ballpark, you're definitely equipped to nail open-ended interview questions at top companies.

The last thing I want to say is that all the above concepts are general guidelines. They're here to help you visualize, organize and train, but they're not law, and they are other ways to think about the topic. Once you're comfortable with the key principles, don't obsess over them and start playing with the rules. That's also what great improvisers do 🎸