I remember back when I was a candidate how much I appreciated good feedback given by other candidates. You know it when you see it: those that point out specifically to what you did wrong, why it wasn’t the best approach and so on. Those where your case partner would suggest a couple of ways to do it differently, down to the level of how to say it. They were the only way I could meaningfully improve, I knew. But man, were they hard to find.
Most feedback wasn’t really useful. Just generic pieces of advice you can read in any book out there, and that apply to pretty much any candidate. “You should be more structured”, or “try to use more logic”, or “try being more insightful next time”. I loathed those. They didn’t help me at all, and worse, got me super insecure with my performance. Would my interviewer think that as well? I wasn’t sure. And this type of feedback didn’t help me improve either, so it was a lose-lose situation.
Take “try being more insightful on your structure”, for instance. What does that even mean? How could I possibly improve that? I mean, if I had more insight I would’ve done it already. Getting that type of feedback would get both my levels of frustration and anxiety skyrocketing.
I knew good feedback was detailed. I also knew insightful structures were those full of specific details. But it was as hard to get the first as to create the second.
I didn’t know it back then, but what I needed was a simple test. A test that allowed me to know if my structure was insightful enough regardless of how thoughtful the feedback I got from my case partner was.
So you don’t have to figure it all out by yourself, I’ve done the heavy-lifting for you. I’ve created a test for you to know how insightful your structure is. Using the Toothbrush Test, you’ll take less than a minute to know that and how to take them to the next level.
The problem with structures that lack insight is that they’re too generic. The Toothbrush Test is a quick measure of how much more specificity you need to put into it.
Most structures err on the side of either not being MECE or not being insightful enough. Those are the two most common mistakes. And although most people understand there are techniques you can use to ensure your structure is MECE, few candidates are aware you can be systematic about being insightful as well.
There is both “art” and “science” to that. The “art” part is the unique ideas and hypotheses you can generate throughout the case that other candidates can’t. This is tough to develop and come from both work and case experience. But the “science” part is as important and much simpler: to have case-specificity built into your structure. Generic structures are NEVER insightful, so your first step towards business insight is to have as many specific (i.e. non-generic) issues in your structure as possible.
But to do it you first need to understand how much of your structure is specific in the first place. The key to that is what we call “The Toothbrush Test”.
The test itself is pretty simple: count how many issues (or “bullet points”) are there in your structure and see how many of them could be used to solve a case in another industry, for instance, a case in the toothbrush industry. Your “Toothbrush Test Score” (TTS) is the % of issues you’ve raised that are specific to your case and wouldn’t function if the case were about selling toothbrushes. The higher the number of specific issues you have, the higher the chances you’re bringing a fair amount of insight in your case.
It’s a simple test, and this is why it’s so effective.
Let me show you how structures with different TTS are different. I’ll start with a structure with an obscenely low score.
This is a typical structure that would get a feedback of “not insightful” or “lacks business sense” in a case interview with a top consulting firm. This candidate WILL be rejected even if he could solve the case. Why? Because you can “solve” any case with it, but not solve any casewell with it.
The problem of this structure has more to do with the individual issues than with the categories. Although both are very generic, you can get away with using generic categories, but the issues have to be specific no matter what.
Except for one issue (of 13), all others would help you solve a case on a toothbrush manufacturer entering the african market. Or a case about a car manufacturer entering Africa. Or a case about an airline entering the asian market. It doesn’t matter. This structure is so generic that if I hide the case question and switch the word “insurance”, you will have NO WAY to tell what topic this case is about.
Consider what I just said for a second. YOU HAVE NO WAY OF TELLING IF THE CASE IS ABOUT INSURANCE OR TOOTHBRUSHES. Those are two completely different industries and you still have no large differences between both structures. That’s incredible, to say the least. Especially considering how many candidates use structures like this on interview day.
With a less than 10% Toothbrush Test Score, you have no chances of surviving a consulting interview. You’ll get rejected no matter what you do next.
Let’s move on to an example of a structure with a better score to see how good it is and what are the implications.
This is a good structure, with an adequate volume of insight. Generally, a TTS of about 40% is enough to show the interviewer you are thinking about this specific case and not blindly using a generic framework memorized from somewhere.
You’re gonna notice this structure for the private jet case is much more specific than the one for the insurance case. And the reason it is specific is that some of the issues are specific enough that they could be used in this case, but not in others. These are the issues that get a YES on the Toothbrush Test. Because they’re roughly 40% of the issues, I gave this structure an approximate score of 40%.
A couple of the issues got a “MEH”, which means they’re almost there. They couldn’t be used for a case in the toothbrush industry specifically, but they could be used for plenty of other cases in other situations related to airlines or the maintenance service industry. I didn’t count these as a YES, but they do make the structure a little better.
The takeaway here is that to have a structure that is specific enough to a case, you don’t need to have a custom-tailored framework, but to have some custom-tailored issues within a fairly generic framework.
To get you the clearest picture possible, let me go through each of the issues that have passed the test so you can see why they have:
- What are the shares of each market segment? (urgent vs. preventive, by type of job, by type of aircraft): Shares of each market segment is pretty generic, but what’s inside the parenthesis (which is something you would have said out loud during the interview, as examples) is specific to this case and shows insight to the interviewer – it shows you’re thinking about this specific problem.
- What are the largest contracts our competitors have with major airlines and what are their terms and conditions?: When you say this you imply to your interviewer that (a) maintenance is sold via contracts, and (b) that their terms and conditions (length, for instance) could make your entry more difficult. These factors don’t apply to toothbrush selling but are highly relevant to aircraft maintenance services.
- What are the barriers to entry in this market? (e.g. getting contracts with major airlines, regulatory permits, being present in several airports): Much like issue number 1, the “barriers to entry” part is pretty generic, but the examples given are specific and to the point.
- Do we have the expertise and the licenses to do airliner maintenance?: Maintaining aircraft requires both specific engineering expertise and hard-to-get licenses from regulators. Making toothbrushes is a much simpler endeavor (although you could talk about engineering expertise for brushes, licenses aren’t as relevant).
- Can we adapt our operations to our customer’s schedules?: Every airline’s nightmare is to have their aircrafts grounded. Poor maintenance scheduling does just that. By raising this issue you imply this understanding, which is pretty specific to this industry.
All of these issues are giving specific examples of how generic things would play out in this industry or implying an understanding of how this industry works. These are a couple of practical ways to make your issues case-specific.
Two of the issues were just a bit short of making the cut (I called them “MEH”). Let’s take a look at why they were better than generic, and yet didn’t pass the test:
- How do they contract this service and what drives their loyalty?: This issue is much better than a generic one for two reasons. First, it shows understanding that this service is contracted out, not bought as a one-off (as toothbrushes are). Second, it implies loyalty is important to this specific industry. So far so good, but the problem is that plenty of industries work on a contract basis and depend on customer loyalty. This specific issue wouldn’t work in a toothbrush case, but it would in several other cases.
- How can we have bargaining power over leader airlines?: Much is implied here, which is good; but little is explicit, which is bad. This issue could be highly relevant for this case because the airline industry is concentrated, so it is possible big players provide little profitability. Again, this issue is specific enough that it wouldn’t work for a case on toothbrushes but generic enough that it would work for plenty of cases.
You want to get as specific as possible with as many issues as possible within your structures.
As a rule of thumb, a 40% TTS is enough to show your interviewer that you’re thinking about this specific case and that you can find key issues within an industry that aren’t obvious. (Obvious = key issues that are key to any industry).
But what if you aimed higher? What if your structure had an even higher TTS?
Let me show you one of those.
This is a structure with a ~90% TTS.
I have added one complication to show a nuance of the Toothbrush Test: I did a case on hair care. Because selling shampoo isn’t very different from selling toothbrushes, some of the issues here could be applied to a case on a high-end toothbrush brand trying to enter the Asian market (the 4th bullet point within Customers is an example of that). Even then, the issue still passes the Toothbrush Test, because it wouldn’t fit well in a case on the Airline industry, or a case on the Mining industry, or a case on the Call Center industry.
What the Toothbrush Test tries to do is to understand how many of the issues are specific to the case. There’s no need to get toothbrush fixation.
I notice three interesting things about this 90% TTS structure.
The first is that it’s obvious to anyone that this candidate has deeply thought about the problem before presenting the structure. It mentions so many things specific to selling high-end shampoo in Asia that this person couldn’t possibly have memorized this or be using stuff they saw in a “similar” case. The structure talks about showers, hair genetics, retailer shelf space, and how affluent Asians look at beauty. Once you start having issues like these in your structures you’ll immediately erase any doubts from your interviewer’s mind that you’re an independent thinker.
The second interesting thing is that the issues are longer. Each bullet point has more words within it. This is logical: more specificity requires more words. But you can’t write that much without wasting a lot of time, can you? Well, you don’t need to. In a real case interview, you can simply jot down a few words for each issue on paper, and be specific when you speak. This means you’ll be talking through each issue meticulously and your written structure will be shorter – jotted down words to remind you about the topics at a glance.
The third thing I wanna point out is that the broad categories are not customized at all. Even though this framework scores 90% of the Toothbrush Test Score, it uses the classic 3C1P framework. I did this on purpose to show that (i) the Test just considers the specificity of the issues, and (ii) you can create pretty good structures using pretty generic frameworks.
This third point also shows an opportunity for you to improve your structures through other means besides making the issues more specific. If your TTS is constantly around 60%, it may be wiser to try to improve your structuring ability through other means besides pushing this score to even higher levels. As we’ve seen, 40% is enough in most cases, so the difference between a 50-60% and a 90% structure is marginal.
Now that you understand the metric, your next step is to find your average TTS.
All this stuff ain’t useful until it’s useful, right?
If you found this concept useful, you might be asking yourself “how the hell do I apply this in practice?”
I have a suggestion that will take you 5-10 minutes and help you determine your next steps in your preparation journey: grab whatever notebook you use to practice your cases and take a look at the last 5 cases you’ve solved. Do more if you feel inspired, but 5 is enough. For each framework/structure you created to start the case, try to find its approximate TTS. You can do the same as I did with the structures in this post (give each issue a YES, MEH or NO, and do the percentage of YES/TOTAL).
If in doubt if an issue is specific enough or not, it probably does not pass the test.
By doing this exercise, you’ll find yourself in one of four scenarios, which I’ve described in the table below.
Would be glad to hear from you in the comments below. What is your average TTS? What steps will you take to improve it? How will it help your interview performance?