The world’s leading publication for data science, AI, and ML professionals.

How to Prepare for Business Case Interview Questions as a Data Scientist

A guide for increasing your business acumen and acing the interview questions

Office Hours

Photo by Hunters Race on Unsplash
Photo by Hunters Race on Unsplash

Why are business case interview questions so important?

It’s not enough to be good at statistical tests, machine learning, or coding. These technical skills are, of course, essential to being good at Data Science. But it’s possible to know all the technical things and still be considered a terrible data scientist. One also needs the soft skills and business knowledge to be able to work effectively with others cross-functionally, communicate results, and really understand the problems you are trying to solve. Having some business acumen is going to make you a much more effective data scientist.

Do you understand how your work connects to the larger whole of the business? Do you understand the overarching goals of the company and how data can support those goals? Can you do work that actually drives actionable change for the company on problems both big and small? These are the things that the business is trying to assess when they ask you business case interview questions.

Now that you are convinced of why business knowledge is important, let’s talk about how to gain such knowledge and how to prepare for the interview questions. It can feel confusing to know how to study for these kinds of questions because they are very open ended and there is no "correct" answer the way that there is for, say, a probability question. If you want to be better at coding questions, you practice your coding. But how do you practice "business"?

1. Build a foundation

Photo by Arnold Dogelis on Unsplash
Photo by Arnold Dogelis on Unsplash

Start by getting some business basics down. Here’s some things to research:

  • Project management methodologies: How do companies organize the getting done of projects? Learn about models like Waterfall and Scrum. What are the advantages and disadvantages of different methods? When might a company use one model over another for different types of projects?
  • Organizational roles: Understanding the different roles at companies will help you better understand how work gets done. Start at the top with executive level roles. Then move down into other typical roles at a company, especially those that a data scientist might work cross-functionally with, like software engineer, product designer, product manager, etc. This can be an especially valuable exercise with a specific company in mind. If you have an interview lined up already, spend time trying to find how their organization is structured and how the different teams work together – especially for your role. Going into an interview with a knowledge of how the company integrates work between different teams is going to help you immensely.
  • Business cases: Learning about a business case and how it is written will help you get in the frame of mind of a business. You may never have to write business cases as part of your job, but it’s not enough to have a great idea with data if you can’t frame it in some kind of sensical way for the business to understand what benefit it will have. Being familiar with how business cases are written will help you understand how businesses make decisions.
  • Metrics and other terminology: Take note of all the acronyms or terms you come across that you don’t know. These most often happen with metrics. Things like ROI, KPI, CTR, conversion rate, and customer churn are very common and come up all the time so know what they mean and how they are used. If you are looking for a job in a certain industry or niche, then spend time looking up different metrics or terms relevant to that industry. For example, something like MAU (monthly active users) is an important metric for companies in a social media sphere, sensitivity and specificity are important for a company that makes medical tests, etc.
  • Tools and technology: Look into common tools for how companies get work done. What options exist for how companies manage projects, version code, or collaborate? I find that looking at specific tools helps to understand actual workflow. This is a great thing to research about a company before an interview.

All of these topics can be researched generally to get a better handle on business knowledge and vocabulary, or they can be done for specific companies. Any time you have an interview, you should go through each of these items for that company as part of your prep. Or you can also research these more generally if your goal is just to get more of a basis for business thinking.

2. Company specifics

Photo by Phil Hearing on Unsplash
Photo by Phil Hearing on Unsplash

Now that you have some basic knowledge, it’s time to dive into more details. From here going forward, it’s best to have a specific company in mind. If you don’t specifically have an interview lined up and are just looking to practice, then think of an ideal company you would like to work for and pretend you are preparing for that interview.

Here is what to research:

  • Company vision and goals: Spend some time getting to know that company. How do they make money? Who are their customers? What are their high level goals and how are they achieving them? How do you think data fits into that equation?
  • Company products: Sometimes it can be weird to identify company products because the most obvious products are physical goods. But a product could be a software, service, app, etc. How do each of these products fit together? Who is the customer for each product? What problem is each product trying to solve?

There is lots of advice out there to research a company before interviewing and I think most (good) candidates follow through on that. However, this point I have outlined so far is often where people stop and they don’t think to keep going.

3. Products, products, products

Photo by Sander Crombach on Unsplash
Photo by Sander Crombach on Unsplash

This is where you are going to separate yourself from the average candidate. This section is going to get more into the nitty gritty that will prepare you for those business case interview questions. Here’s the process:

  • Start with a company product: There was some studying of the products in the previous section, but look at each specific product in depth. With each product, remind yourself of how it works, the customer, and the problem it is solving. Now think about things like what features it has, what drawbacks or problems might there be with it, and what kind of data you think the company collects related to the product.
  • Imagine you are on the team for developing this product: Put yourself in the shoes of the people working on the product before it existed. What kind of problems do you think they faced? What kind of data did they collect? How did they measure success? What kind of metrics do you think they looked at as guardrail metrics? What kind of statistical experiments do you think they used?
  • Imagine you are on the team for improving this product: What sort of improvements have been made to the product over time? Think about how you would identify needs using data. What kind of data would you collect? What would be your key metrics for success? If you were to think of a future improvement on this product, what would it be? What kind of experiments would you design to test it?

Repeat this process for each of the different products. The more you can know the ins and outs of products and metrics at the company, the more prepared you will be to answer business case questions. Mentally putting yourself in the role of Data Scientist at the company is going to push you to really envision the needs and struggles of the business and how your work would solve them.

4. Structuring your answers

Photo by conner bowe on Unsplash
Photo by conner bowe on Unsplash

Structuring your answers can feel like the trickiest part of answering these questions well. The most important thing to do is to let go of the idea of there being a "right" way to do this. You should treat business case questions more like a conversation which can make it difficult to keep to a set structure. In this way they are more similar to the HR type questions where they are open ended and really just depend on the situation, question, company, and your experience. But just like HR type questions, the more you practice then the easier it will get to coherently talk through answers in a logical way.

I like to borrow ideas from how a business case is put together to help me think through the parts of my answers. I recommend looking up typical questions and answering them to yourself in the mirror or get a friend to act as your interviewer. Here is a general structure that I use to help me begin thinking more linearly about these problems. I hope that this will help you begin to organize your answers in a way that makes sense to you:

  • Clarify the problem: Like any data project, you should always start with a clear objective or problem to solve in mind. Let’s say you get asked something like, "XYZ product is 20% down in sales, what do you do?" An important place to start is to clarify what "20% down" means. Does that mean sales compared to this time last year? Since last month? Is the drop measured by raw dollar amount or by number sold? This clarification will help you identify exactly what the problem is and what metrics/data to talk about. Remember that you should already be familiar with the common metrics used at this company and what their products are. You want to portray that you can approach problems logically and define what needs solving. In real life, this is the most important step as it will cause the most pain if you aren’t on the right page.
  • Identify possible causes and solutions: Now that you know what the problem is and what metrics are being targeted you can think through some possible causes. Companies often want to see that you can come up with multiple ideas so I would suggest you aim for 2–3 ideas so that you are showing your ability to scope a problem while also keeping your answer organized. Again, think of what the company is looking for in the way you work – they want someone who can bring ideas to the table.
  • Talk about experimenting: How might you set up an experiment to test your possible causes/solutions? How would you sample? What sort of hypotheses will you come up with if you see metrics increasing or decreasing? When might increasing or decreasing metrics be deceiving? Sometimes you may have interviewers that want you to get more in the weeds or stay at a high level and that is something you are just going to have to assess from the conversation. But generally, the company is trying to see how you would take action in applying your data science knowledge to a real world problem.
  • Identify pros and cons: How might some metrics affect each other? How expensive and involved might solutions be? What would be the risks and benefits of different solutions? What kind of recommendations might you make in this hypothetical scenario? The interviewer wants to judge how you would communicate results, assess short-comings, and give recommendations.

Really utilize your interviewer to have a discussion. If they aren’t giving you much feedback about if you are on the right track, then you are completely allowed to ask if you are or not. If you are worried you are rambling too much, then ask if they would like you to keep going down a certain track or to move on. Keep in mind that just like all the other interview questions, the company is trying to see what it would be like to work with you. Always answer questions with the guiding principle in your mind of displaying how you would add value.

Conclusion

Hopefully this guide gives you some good basics to work off of and expand upon. Remember that there are no right answers to business case questions, it’s about both parties trying to see if you are the right fit for the job. For that reason, you should always strive for your answers to put your best foot forward in displaying the way that you think and how it would be to work with you.

Good luck on those interviews. You’ll do great!


Related Articles