In the McKinsey interview you will have to answer three different questions types - broadly speaking:
- Exhibit Interpretation
Now for Structure and Exhibit Interpretation, there is no right or wrong answer. Some answers are better than others because they are
- follow a strong communication (MECE, top-down, signposted)
That being said, there is no 100% that you can reach or the one-and-only solution/ answer. It is important that your answers display the characteristics specified above and supported well with arguments.
As for Math questions, usually, there are answers which are correct (not always 100% the same since some candidates simplify or round differently - which is ok), and others that are wrong, either due to the
- calculation approach
- calculation itself
Now, for the interviewer, the overall picture counts. Mistakes in one area need to be balanced by a strong performance in other areas. McKinsey wants to see spikes in performance in certain areas and a good enough performance in other areas.
The most common example I see almost every day: You can be strong in structure and exhibit, yet make a small mistake in the math section - overall as you might consider 80% - and still pass on to the next round.
On the matter of industry-specific cases:
It will be a mix of cases in a domain-relevant context as well as cases set in a completely different context.
Be aware that frameworks were applicable in the 2000 years, the era of Victor Cheng and Case in Point. McK has long caught up on this and the cases you will get during the interviews are tailored in a way to test your creativity and ability to generate insights, not remember specific frameworks.
In fact, it will hurt you when you try to use a framework on a case that calls for a completely different approach. Also, it gives a false sense of security that will translate to stress once you figure out how your approach won't work during the real interview - I have seen this so many times...
Your goal should be to learn how to build issue trees, interpret charts, and perform math no matter the context, industry, or function of the case. I have a system for that, especially for McKinsey cases, that teaches you exactly this approach + a ton of exercises I give my coachees to progress their ability to come up with deep, broad, and insightful answers for each case individually.