Case interview theory with Anna Saakyan by Fless

Video 1. Cases with Anna Saakyan. Formula structures.

Cases with Anna Saakyan. Formulas as the base of case structures.

Hi, everyone! My name is Anna Saakyan and we are launching a new show “Cases with Anna Saakyan”. In this show I’ll talk about some lifehacks for solving cases and passing case interviews. We’ll also invite guests as candidates or representatives of the companies whom we’ll ask to act as interviewers.

To begin with, let’s find out what a case is and why it should be solved structurally. A case is a small task or a problem we are usually asked to solve when we want to enter a consulting company. But let me note that it will be useful not only for those who want to work in consulting but also for those other people who deal with business tasks in their everyday life.

While solving a case we can just guess the answer – this is what many of those who do not practice case solving do. Or we can give a non-structured solution, i.e. brainstorm. The main idea of our approach is to demonstrate structure and come up with a certain solution based on this structure.

In general, all the cases can be divided into two groups – cases aimed at finding a problem and cases aimed at finding a potential. It is very important because it influences the way we run the case and the questions we ask along the way. But no matter which type of case you use they will always have the same structure.

Let’s talk more in detail about structures. We distinguish three types of structures that usually intersect with each other. They are formulaqualitative structureand process structure. Why are they shown that way? Because it seems to me that a different graphical representation allows us to immediately determine which type of structure is in front of us and makes the perception of our decision easier.

So, in the squares I have the structure according to the formula, in the circles – the structuring of quality parameters, which is called a qualitative issue tree, and we also draw the process using arrows. The process can be, for example, a value chain or a customer journey description.

Let’s talk about the characteristics of these structures. What is the meaning of each structure and what are the differences between them?

The formula is based on a numerical indicator, and our task is to decompose this indicator in order to break it into smaller drivers. Accordingly, when we are looking for a problem or potential for improving an indicator, we decompose it and use the formula for this.

A qualitative structure is either a solution or some quality indicator. For example, a decision tree is a qualitative structure when we group certain types of solutions into different blocks. Also, the qualitative structure can be used to structure various factors or parameters that determine a particular characteristic. For example, we can structure in blocks the parameters that characterize a product, i.e. the technical side of the product, the external side of the product, the perception of the product, etc. Why we draw it like this? Because it is easier to perceive different types of structures in different visualizations.

The process characterizes either the value chain or the customer journey – customer’s track to obtaining this value.

Any of these three structures is usually used in combination with another structure, and this gives us a complete set of tools that allow us to either find the problem or find the potential for improving this or that indicator.

Let’s take a simple example to explain how the structural approach differs from the non-structural one. Let’s make an analysis based on the example of the problem search. Finding a problem is, in fact, an easier case than searching for potential. Because usually we just need to find one or more driver that has changed, while finding the potential is more difficult, and there may be much more drivers involved in the case.

Accordingly, let’s take, for instance, Starbucks and imagine that Starbucks’s revenue has fallen. We take not the entire network, but a separate location, a particular cafe in a particular place. So, we need to understand why the Starbucks’s revenue has fallen.

What does a non-structural approach mean? It is the approach of just trying to guess and to give the most common, in our opinion, reasons why the café’s revenue fell. For example, maybe a competitor suddenly appeared. Yes, this is probably one of the reasons – maybe there is a competitor, so now there are fewer customers coming to our place. The second reason – maybe people have less money, i.e. no money, and in this case, probably, people stopped going not only to Starbucks, but also to all other cafes. Maybe people have ceased to like our product, and they now go to other places that offer them something else – accordingly, there may be a problem in the product. Or maybe we removed the signboard, and now people don’t know – people who didn’t go to our cafe before, don’t know how to find us.

If we continue our brainstorm, we will find a big number of such possible drivers. But the problem is that they are not structured. So, how is it possible to structure the problem with the formula approach? 

We know that revenue is a certain numerical indicator. Accordingly, it can be decomposed into other numerical indicators each of which has a certain impact on the total one. We know that revenue is measured in money. So, this is the main indicator that has decreased. Next, we should understand what affects the formation of revenue in the café. As far as we know, it is the number of customers multiplied by the average bill. Starbucks is a cafe in which many people take coffee with them, so the cafe’s traffic greatly influences the number of customers. For now, we’ll just write the number of customers and the average bill from a customer. This is our first level of hypotheses. What does this mean – it means that we understand that revenue could have changed either because there were fewer customers, or because the average bill decreased, or because both indicators had somehow changed. 

We will continue to decompose these indicators according to the formula. For example, we were told that the average bill had changed. This means that we need to understand exactly which driver, that forms the average bill, had led to a change. The average bill can be decomposed into the number of positions in the bill and the average cost of one position. After we find out which of these indicators had changed, we can understand what happened – whether there was a price reduction, or there was a decrease in the number of positions that customers bought.

From here, I would like to show the passage from formula to qualitative structure. So, if the average cost of a position has decreased – it can change either if we have reduced prices, or if we began to provide more discounts, or because the product mix has changed. The product mix in this case will mean the ratio of positions with different prices in our bill. So, this is a qualitative structure, which already gives us more precise reasons of the change of this quantitative parameter.

Let’s go further. If the number of positions in the bill changed, then first of all it could have happened either because we had changed the assortment, i.e. we began to offer fewer positions or less interesting positions, or because the percentage of sold positions from this range became smaller. This can be driven further either by the preferences of people which might have changed, or by our sales. I.e., for example, previously we offered additional syrup to the coffee, then we ceased to offer it, and therefore we now have less sales of syrup. It is the second reason that can be considered here.

Let’s go ahead. If the number of customers who come to our cafe changed, it, as I said, can be driven by the flow. In fact, I would consider the number of customers of a cafe, shop or restaurant as the indicators that depends either on the flow or the people living in the area. This difference will be considered depending on the area in which Starbucks is situated. For example, if we take Starbucks – Starbucks is usually located in the center, and it is intended for people passing by or working somewhere, i.e. it is not the type of cafe located in residential areas. Accordingly, here I would recommend taking the flow of people who pass, and the conversion from this flow to our clients. We have fewer customers, it means that either the flow has decreased – the flow is also a number, we can also calculate it – or the conversion, i.e. the percentage of people who visit us from this flow has decreased.

Accordingly, further on the basis of these two parameters, we can also look at qualitative reasons. For example, what drives a stream in a certain area? Most likely, this flow is driven by the presence of some anchor buildings or constructions, i.e. either a shopping center, or a road junction, or something else. So, first of all, here we will make an oval, because we consider the qualitative structure. So, the flow could have changed either because the ratio of some anchor positions in this region changed – i.e. the shopping center or business center was removed, or because the transport situation somehow changed, the subway exit was moved to another place. Or, in general, the ratio of people has changed.

Here you can further make a more detailed analysis, depending on which particular area we will consider and what data we will receive from the interviewer.

On the other hand, if the conversion has changed, then this may be either due to the fact that we have competitors, or due to the fact that the customers’ preferences have changed in some parameters, or because we changed something in our offer. And all this can also be considered by a number of qualitative characteristics, which can be used to evaluate the Starbucks offer.

As a result, from our problem we derived a rather branched structure, which combines both formula indicators and qualitative indicators.

In this case we do not come to any specific problem, because our task was to examine all the blocks in as much detail as possible. However, if we were solving a real case, then the problem could be in each of these blocks. Here we have shown in general a path how one could come to this or that problem.

In the following videos we will tell you in detail about the use of the qualitative structure, and also describe how you can use the quantitative and qualitative structures in the description of processes. Follow us for our next releases. Good luck. 

Next video.