Guide To Data Analysis & Interpretation in Case Interviews

A common type of question posed in case interviews is data interpretation and analysis.

What typically happens is, during the framework/hypothesis phase, you talk about your initial thoughts and ask for more data on 'XYZ' (re: in a profitability case, you may want to dig deeper into revenues and ask for pricing/volume data, in a market entry case you may want to dig deeper into market size and shares, etc.).

Upon asking for more data, the interviewer may smile and proceed to hand you some graphs. Regardless of case style - interviewee or interviewer led - analyzing data, graphs, and tables is a staple in the case interview process.

There are three forms in which data can be presented to you in a case interview:

1. Numerical data in a table form.

2. Bar / pie charts w/ axes that indicate units and reference data points

3. A mixture of tabular and graphical representation of data

You will be tested and evaluated on the following:

1. Analysis: Quickly understand the information presented in the exhibit. Depending on the case, you may be tasked with finding: what overall market size is, how much did revenues grow in the last year, which customer segments saw the greatest decreases in revenues, etc.

2. Contextualization: Understand how this data fits with the overall context of the case, as well as the specific context of the current discussion / section of the case. Things to consider, depending on the case and situation, may include: is the market big enough, given our client’s existing operations? What do we know about the client that could explain the growth in revenue? Is the new competing product, which you previously discussed, potentially responsible for slowing down growth in a particular customer segment?

3. Interpretation: Present your business insights, and continue the case in the right direction by asking relevant questions. Perhaps the data presented suggests that the new market is 5x larger than the client’s current market and faces fragmented competition — therefore, your hypothesis may be that this is a good market for your client to enter. In another data analysis question, you may see how certain products have disproportionately positively contributed to revenues, or that a new product launched by your client’s competitors is mostly responsible for decreasing your client’s market share in Segment X of customers.

As you analyze the data, here are some best practices:

Keep in mind the question you were asked when the interviewer handed you the exhibit. Also keep in mind the main question and objective, as stated at the very beginning of the case. You must be prepared to understand what the data is telling you, but also tie it back to the original question. What is driving your client’s unprofitability? Is this a good market / should we enter? Should we acquire target company ABC? How does the data help you answer a specific question, and then how does it tie back to the original prompt?

Read the title of the exhibit, and then study the axes and units of the data (if examining charts). Reading the title will help buy you some time, and explaining what the axes mean in relation to the data will again help you understand what the data is trying to tell you.

Example: As we can see, this exhibit is a bar chart that outlines Consumer Preferences for healthy non-alcholic beverages, sports drinks, and leisure beverages. In particular, this chart elucidates how consumers identify our new product in relation to our competitors Cool and O2Flavour. While ‘Cool’ is associated with health non-alcoholic beverages and ‘O2Flavour’ is associated with leisure beverages, consumers don’t identify them in the sports drink category. When considering launching our product, we should think about positioning ourselves as a sports drink, and appropriately target distribution channels such as sports events and athletic venues, and even partner with athletes to promote our products. Our next steps should be to discuss how we can position ourselves to best embody the sports drink identity, and how - from a marketing and promotion perspective - we can help our client best do so. What do you think?

Figure out the most efficient approach to get to the final result. Of course, this depends on what graph you’re given, but in general it pays dividends to look at the delta, or change, in growth (positive or negative) from beginning to end.

Example: As we can see, this exhibit is a bar chart that outlines Chris’ clothing sales and costs (revenues and expenses). In particular, the data elucidates how revenues and expenses have changed over the years. At first glance, it seems both have increased over the years. But is revenues increasing faster than costs? Let’s find out from Chris’ ‘retail’ perspective (wholesale and website sales are negligible, focus on what’s efficient, remember) — from 2011 to 2015, sales have increased from $500M to $550M, a 10% increase. In the same time frame, costs have increased $250M to $300M, a 20% increase. In other words, costs are increasing twice as fast as sales. This could be occurring for a number of reasons: variable costs such as COGS and raw materials (think fabrics, wool, cotton, etc.) might have increased over the years, labor or distribution have become more expensive, or fixed costs like expensive marketing campaigns, social media managers’ salaries, and general costs of administration have ballooned in the past 5 years. Our next steps should be to dig deeper into the expense side of the profitability equation, and identify key drivers of increasing costs. Do we have any data on how fixed and/or variable costs have changed over the past 5 years?

Interpreting and communicating your insights to the interviewer

The final and most important step is synthesizing all the analysis and business reasoning that you just completed, and presenting it to the interviewer in an effective manner. The following aspects should be taken care of while communicating the inferences to the interviewer:

1. Hypothesis: Consulting is hypothesis-driven, and you should try to lead with a preliminary conclusion, backed by information you currently have, and can be verified by asking for more information. Ask for the further information needed to verify your hypothesis.

2. Data: If you've been given numbers and made mathematical calculations, it makes sense to present the most important results to the interviewer, to show your comfort with numbers. Walk them through any calculations you have conducted --- especially in virtual interviews, where you can't easily share your notes, you must make sure you don't leave your interviewer in the dust, so to speak, and ensure they understand your math + thought processes.

3. Clarity & Logic: The communication needs to be precise and logical. You have to explain the logic behind your conclusions. In general, it often helps to think aloud, so that the interviewer can follow your line of reasoning.

You’ve seen these best practices play out in the examples above, so when casing be sure to employ these tips and tricks to stand out in your interview!