Right now I'm feeling like I don't quite understand consulting... what do consultants think of themselves? What do clients expect out of consultants?.. For my junior level consultant friends, I know they were just involved in brainstorming, and did all the cold call, data mining and modeling and etc..
The very first time I heard about a case interview, my first thought was, "Who in the world invented this torture device?"
After going through the interview process, becoming an interviewer myself, and observing the more recent innovations in case interviewing (e.g., group format, data sheets, etc...), I have a different perspective on the matter.
In preparing for case interviews, whether you have an McKinsey interview or a bain interview, it's often very easy to lose sight of why case interviews exist in the first place. Their role is NOT to torture applicants arbitrarily to see who can jump through more hoops.
In fact, the case interview process (and all its incarnations) do a surprisingly good job at mirroring the on-the-job experience of being a consultant.
Yes the purpose of the case interview is to find good consultants... easy to forget it isn't it?
So there are a few implications:
1) The case interview is surprisingly similar to the on-the-job experience of being a consultant. If you hate case interviews, you will very likely hate the job.
2) The reason you get "estimation questions" (e.g., estimate the market size of the used car auto sales market in the United States) is because clients ask consultants these questions all the time. Client gets a random idea, hey why don't we consider getting into the used car sales market? I wonder how big the market is? Hey, Victor (or your name) any idea how big this market is?
"Well, lets see... there are 350 million people in America...."
I certainly did estimation questions during the interview process.... but once I got the job, I ended up doing MORE estimation questions during my work as a consultant, than I did interviewing to become one.
3) In some countries outside the US, I am hearing that some firms are having very short 5 minute, pre-first round interviews. My best guess is these firms are trying to evaluate language fluency and business maturity - both of which are hard to assess via a resume.
In addition, I have heard of some applicants getting off the wall questions that aren't standard case interview questions, aren't consulting resume questions, and are basically impossible to answer in 3 minutes.
If I had to guess, I think this kind of question is a composure test. The answer you give is less important than how you give it. The interviewer is looking to see if you remain composed.
Do you make up a BS answer and try to fake it (a bad idea), do you happen to have a intelligent answer and can you back it up, or do you have no clue and can confidently say, you have no idea?
The 2nd or 3rd answer is acceptable, the first is not. Never like or make up an answer because on the job you would never want to lie to a client or BS a client... because eventually they find out and it reflects poorly on the firm's reputation. So in this particular answer, the interviewer is trying to see if you're the kind of person who has sufficient confidence and judgment to say you don't know the answer to the question.
Again, on-the-job it's impossible to know everything. I have had clients ask me question to which I did not know the answer or did not have the data to backup an answer. I routinely would tell clients I didn't know (but could find out), could do a quick estimation to see if the answer was even relevant, or could venture a few guesses (hypotheses) but would say that I couldn't support them yet.
4) Basically everything that happens in a case interview happens because it simulates some aspect of the on-the-job experience. So when you are asked a random question, stop thinking like a candidate trying to impress the interviewer. Think like a consultant and ask yourself what would you feel comfortable saying to a client -- knowing your firms' reputation is on the line (because it is).
Here's a bit more color on what it's like to be a consultant.
When you start with a new client, often there is skepticism about the value the consulting firm can bring (and in some cases a particular consultant... like a really young one, or one without experience in the client's industry).
In short, you (or your firm) has to prove yourself. Your opinions don't count for much (because everyone has them and you typically have less experience in the client's industry than the client does).
So all you have to go on to get taken seriously is 1) asking interesting questions client's don't know the answer to, 2) getting or analyzing data in a way that brings out new insights that client hasn't seen before, 3) coming up with data-supported conclusion (especially data-supported conclusions that are counter-intuitive) that lead the client towards a different set of decisions.
Often the client or certain members of the client are looking to discredit you... so they can get back to their real jobs of running the company, instead of baby-sitting a consultant. So a semi-hostile client is looking for you to screw up somehow.
The two most common screws ups are: 1) inter-personal relationship screwups (being rude, arrogant, or dismissive), 2) saying something you can't support with data and communicating that something as if you had data.
So early in a consulting firm's relationship with a client, the consulting team is looking to prove itself, bring value through data and analysis, and to use people skills to build relationships with a sometimes hostile client organization.
It's not good enough to be smart analytically, you need to be smart with people too. And you'll notice the common theme amongst the more recent changes in the case interview approach are often designed to see how a candidate's people skills.
As I mentioned before, every aspect of the interview process happens for a reason -- and most often that reason is designed to simulate some aspect of the on-the-job experience.