Definitive guide to prepare for an analytics interview

Looking to land your first data science role but struggling to clear interviews? We have curated the most comprehensive course to ace your next data science interview. Hundreds of questions, plenty of videos, and multiple resources – the ultimate course!

Let’s face it! Facing an analytics interview can be daunting at times!

I have met a lot of analysts, who are good analysts when you interact with them informally. But something happens to them, as soon as they enter into an interview!

Have you seen one of these analysts and wondered what happens to them in the room? Or have you faced this situation yourself? This guide is meant to help you / your friend to ace the next analytics interview!

The first thing to keep in mind before appearing analytics interview is:

As long as you know your subject, are a logical person and can stay calm – you can ace these interviews easily!

What is the employer trying to judge you on?

The actual skills, which the employer might be judging on, would vary from employer to employer, but it is likely a mix of the following skills:

• Technical skills – comfort and knowledge about various analytical tools
• Knowledge of statistics – whether you apply algorithms blindly or actually understand what they do?
• Structured thinking – Can you take ambiguous problems and put a framework around them?
• Business understanding – How well can you put on your business thinking hat?
• Problem solving – Can you provide (out of the box) solutions to problems?
• Communication skills – Can you communicate your thoughts clearly and crisply? Can you influence people?
• Comfort with numbers – How good are you at crunching them?
• Attention to details – Do you pay attention to small details and at them up to see the bigger picture

This article can help you understand the perspective of an employer in some more details.

Types of analytics interviews:

Analytics interviews can be divided in broadly three categories:

The preparation for technical analytics interviews happens over time. These interviews test how much time and efforts have you put, in learning your subject and tools.

If you are really good at what you do, these rounds should be a cake walk. If you are not, the best strategy is to be honest about what you know and what you don’t and let your potential employer know. Here are a few articles to help you with technical interviews:

1. Tricky questions on SAS – part I and part II
2. Tricky questions in R

There is a lot of material available on the internet to prepare for behavioural interview, hence I would skip those details.

Skill assessment interviews:

These are the deciding factor in most of the analytical hiring, and for a good reason – if a person has sound logical skills and can demonstrate good business thinking and logical skills – he can pick up technical skills easily! Since these interviews are aimed to assess various skills, what matters more, is that you demonstrate those skills. The actual answer and solution is irrelevant in most cases. Any hiring manager would prefer a wrong answer with a better approach rather than an accurate answer with bad approach.

Skill interviews, again can be categorized in  2 categories:

1. Guess estimates
2. Case studies and role plays

Guess-estimates:

Guess estimates are puzzle like questions, where you are expected to estimate a figure by putting a framework to a question, creating segments, making assumptions and adding up the numbers to arrive at a number.

You can read details on how to ace a guess-estimate along with a few examples here. Here are a few tips I would recommend:

• It is the approach that matters – not the exact numbers. However, you should cross-validate numbers once you have them.
• Always go top down to solve a problem. Draw neat segmentation & diagrams to illustrate your approach.
• Keep a few common starting points / proxies on your finger tips. Population of your country, population across the globe, the GDP of your country are a few good starting points you should definitely remember.
• Analyze all possible uses of the subject. E.g. You should consider B2B & B2C markets, if you are asked to estimate market of tablets or smartphones.
• Call out assumptions and possible blind spots.

Case studies / Role plays:

Here is what Capital One says about case studies on its website:

Case interviews are broad, two-way discussions rather than one-way tests. You will be assessed more on how you go about dealing with the problem rather than on the specific answers you come up with.

A case typically starts with a broad question providing a business scenario and then narrows down in a particular direction. Cases might also evolve and grow in complexity as the interview progresses. Here is how a typical interview evolves over time:

Here is an example of a typical case study interview. Here is another one.

Following are some best practices to follow in a case study round:

• Case study is all about illustrating 3 things – Structure, structure and structure! Focus on putting framework to the problem provided, and you will be safe. Try deviating from it and you’ll find yourself in trouble.

For example, when asked how can you increase Profits for a product company, you should not jump to conclusions like “I’ll improve marketing or I’ll cut costs”. You should say Profits = Revenues – Costs. In order to increase profits, we can either increase Revenues or reduce costs. Revenues can be increased by increasing Sales or increasing the price. Costs can be reduced by doing ….

Keeping a structure will not only help the interviewer understand you better, it will also help you make sure that you have not missed out any thing.

• Call out assumptions, whenever you are making them. These could be assumptions about business or the sector in discussion.
• Lay out things neatly on paper, such that, they can be re-used later. Most of the times, case studies evolve over time. You will be asked to do similar questions, multiple times under multiple scenario. Keeping them handy can reduce calculation time!
• Think out loud – it is the thinking process, which matters. If you are not sure – ask the interviewer rather than staying quite!
• Communicate crisply and clearly – if you are not clear about your thoughts, take 2 minutes from the interviewer to arrange your thoughts and then communicate them nicely.

Finally, here is a list of activities / behaviour, you should avoid during the interview. These, along with the best practices mentioned above, should give you enough ammunition to handle any analytics interview.

What do you think about this guide? Do you have handy tips and tricks, which helped you in your interviews? Please share your thoughts and practices through comments below. I am looking forward to hear them.

If you like what you just read & want to continue your analytics learning, subscribe to our emails, follow us on twitter or like our facebook page.

image credit: Oregon State University

Kunal Jain 26 Feb 2019

Kunal is a post graduate from IIT Bombay in Aerospace Engineering. He has spent more than 10 years in field of Data Science. His work experience ranges from mature markets like UK to a developing market like India. During this period he has lead teams of various sizes and has worked on various tools like SAS, SPSS, Qlikview, R, Python and Matlab.