Case Interview Fundamentals: Lesson 33 [Drill: Crowded bank branches] — Transcript
Full transcript
- 0:00Hey guys, how are you doing?
- 0:02Um, in today's drill, we're going to see
- 0:05several nuances
- 0:07and it's very specific to advanced
- 0:11candidates. And that's the reason why
- 0:14this is one of my favorites cuz I love
- 0:15the nuances and I love the advanced
- 0:17stuff
- 0:18and teaching it to advanced candidates.
- 0:20All right? So,
- 0:22the case question here is a large retail
- 0:25bank in the US has hired you to help
- 0:27them deal with a problem they're having.
- 0:30They believe their branches are too
- 0:31crowded.
- 0:33So, what's weird here is there is no
- 0:36case question.
- 0:37There's a problem, but there is no
- 0:39question. The The interviewer doesn't
- 0:41even ask you what you would do. He just
- 0:43sits there. So, obviously the first step
- 0:46here is to define the case question.
- 0:49So, think for a while how you would do
- 0:52that, what questions you would ask, and
- 0:55then watch me do it. And then you can
- 0:57compare the answers and see if you would
- 0:59be able to define the case question
- 1:01here, too.
- 1:03But before I do that, I want to comment
- 1:05on something else.
- 1:06So,
- 1:08some of you might think, "Well, this is
- 1:10weird. This looks like a framework
- 1:12question more than like a hypothesis
- 1:15testing question because it starts with
- 1:17the case or because it's not a follow-up
- 1:18question or something like that." And
- 1:20some of you might think, "No, this is a
- 1:22hypothesis testing question because it's
- 1:24very, very specific."
- 1:26And the truth is there is a very large
- 1:30gray area between framework questions
- 1:32and hypothesis testing questions. And
- 1:34many of the actual MBB questions are
- 1:37going to fall into that gray area
- 1:39because if the question is not very
- 1:42specific, then you can always use a
- 1:44framework. So, they always try to bring
- 1:47it either into the gray area or into the
- 1:49hypothesis testing area. All right? And
- 1:52it actually doesn't matter whether it's
- 1:55a framework question or a hypothesis
- 1:57testing question. Because in the end,
- 2:00both have to follow a principle which is
- 2:03the the answer has to be a plan to solve
- 2:06a problem in real life, like you would
- 2:09do in a consulting problem. In a
- 2:10consulting in a consulting project. And
- 2:13interviewer consultants never stop to
- 2:16think, "Well, is this a framework or is
- 2:18this a hypothesis testing
- 2:20problem?" Cuz it doesn't matter to them.
- 2:22It What matters to them is they have to
- 2:23solve that problem. And they don't even
- 2:25use this terminology. We created it.
- 2:27Okay?
- 2:28So,
- 2:30this is how I would
- 2:32define this question. So,
- 2:35the problem definition here.
- 2:37One, why is this a problem? So,
- 2:40uh is it because customers are
- 2:41unsatisfied? Is it because the churn
- 2:43rate is increasing? Why is this a
- 2:45problem? Why is this affecting us? And
- 2:47why do you want it solved?
- 2:48And next, what does the client want my
- 2:51help with? So,
- 2:53uh do they want me to increase customer
- 2:55satisfaction with those services? Do
- 2:57they want me to decrease the churn rate?
- 3:00Do they want me to get people out of the
- 3:01branches? Or do they want me to only
- 3:04figure out why branches are so crowded?
- 3:07So, after getting those answers, I would
- 3:10say, "All right. So, I was hired to help
- 3:12my client figure out if way too crowded
- 3:15branches are driving customer
- 3:16satisfaction down. And if it is, they
- 3:19want me to bring it back to normal
- 3:20levels. Is that correct?" And then the
- 3:23interviewer would say, "You are correct.
- 3:25You can move on." So, again, the first
- 3:28thing I want to do is define my problem
- 3:31and define my
- 3:33hypothesis into a cause, an effect, and
- 3:36a consequence. Okay? So, the potential
- 3:39causes for this is branch employees
- 3:42became inefficient, services that did
- 3:44not require going to the branch suddenly
- 3:46require that, or some issue common to
- 3:49many clients made them go to the
- 3:51branches. There are also other things
- 3:54that could be causing this, like uh
- 3:57a sudden reduction in branch size. So,
- 4:01like maybe they needed to put an area
- 4:03that's not accessible to customers in
- 4:06there, and so the space got reduced and
- 4:09branches are more crowded or something
- 4:11like that. Uh we are going to do a
- 4:15little quick structured brainstorming
- 4:16for this when we're actually structuring
- 4:18our answer.
- 4:20Now,
- 4:21the effect is branches are too crowded,
- 4:23and the consequence is customers are
- 4:26thus unsatisfied with the crowded branch
- 4:29and possibly with having to go to the
- 4:31branch if they if we did something wrong
- 4:34that made them have to go to the branch
- 4:36and that therefore the branch are
- 4:38crowded. Okay? Uh
- 4:40so, the way I would do this, and just to
- 4:44just to be clear here,
- 4:45you could and you could not show this
- 4:49part to the interviewer. I do this
- 4:50because it makes my thoughts clearer.
- 4:53Uh it's fine if you don't show it to the
- 4:55interviewer.
- 4:56Uh I probably would not,
- 4:58uh but it would be okay.
- 4:59So, uh the questions that I want to ask
- 5:03here
- 5:04are
- 5:05one, are the branches really too
- 5:07crowded? Two, are customers unsatisfied
- 5:10with crowded branches and or with having
- 5:12to go there? So, are they too crowded?
- 5:15Are clients upset that they're too
- 5:16crowded? Uh
- 5:18why are the branches suddenly more
- 5:20crowded? So, what's going on? And what
- 5:22can we improve How can we improve
- 5:24customer satisfaction given this? And
- 5:26the reason this last question is
- 5:27structured like that is that it's not
- 5:30necessarily uh
- 5:33getting people out of the agencies, but
- 5:35it's probably uh
- 5:37that way. So, to know if branches are
- 5:41really too crowded, I need to see people
- 5:44per day per square meter, but only
- 5:47divided by those square meters accessed
- 5:49by customers.
- 5:51And that versus competition and over
- 5:53time.
- 5:54And if possible, I'd like to check that
- 5:57per hour to see if there are peaks that
- 6:00are
- 6:01uh too crowded. Maybe it's not in the
- 6:04total number of people per day, but
- 6:07maybe in
- 6:09in
- 6:11the some peaks. Okay? And I'd like to
- 6:13see how that evolves over time to see if
- 6:17it was always the problem or if it just
- 6:19became a problem right now. We could
- 6:21probably easy easily count people in our
- 6:24branches with our
- 6:27employees, maybe even with security, and
- 6:29we'd probably need to get someone in
- 6:33competitors agencies in competitors
- 6:34branches
- 6:35uh to count people in there. Okay? We
- 6:38just choose a few and do that by sample.
- 6:43Next,
- 6:44to find out if customers are
- 6:45unsatisfied, I would do customer
- 6:48research regarding experience in branch
- 6:50versus competition. So I I could get my
- 6:53employees to ask customers how they feel
- 6:55about that.
- 6:57Uh I'd probably start with uh
- 6:59qualitative research to understand
- 7:01what's going on and then move on to
- 7:05uh
- 7:06and
- 7:07net promoter score or something like
- 7:08that to see if customers are happy or
- 7:10unhappy with that. And I'd compare it
- 7:12with competition by getting
- 7:14uh employ uh by putting people doing
- 7:16research outside of competition
- 7:18branches.
- 7:19And if possible, I'd like to check this
- 7:23uh
- 7:24against
- 7:25these results before the problem started
- 7:28if
- 7:29we do this research uh on a regular
- 7:32basis. So maybe we're already
- 7:35uh
- 7:36researching experience in branch
- 7:39uh satisfaction with the experience in
- 7:40branch.
- 7:41Uh that would be great. If that's not
- 7:43possible, that's okay.
- 7:45Uh next,
- 7:47why branches are suddenly more crowded?
- 7:48So, I check if employees suddenly became
- 7:52less efficient, and the way I do that is
- 7:54I would
- 7:56uh
- 7:58try to figure out
- 8:00how many processes or how many services
- 8:05branches are doing per employee per day,
- 8:08and I'd compare that with uh
- 8:10before this problem started. So, it
- 8:13could be that we are now the mix of
- 8:16services changed, and now we're doing
- 8:17more complex uh services. It could be
- 8:20that we changed our employees, and now
- 8:22they're more inefficient. Or it could be
- 8:24that we changed something like the
- 8:26uh software or something, and then
- 8:28they're inefficient because of that. Uh
- 8:30many things could have happened, but
- 8:32first I want to see if they are indeed
- 8:34less efficient.
- 8:35Okay? And next, I try to see if services
- 8:38that did not require to the branch
- 8:40suddenly require that. So, maybe you
- 8:43didn't need to go to the branch to
- 8:46uh
- 8:48I don't know, upgrade your account or
- 8:49something like that, but now you do, and
- 8:51that's why that's there's so many people
- 8:53there.
- 8:54Or maybe there's some issue common to
- 8:55many clients that made them go to the
- 8:57branches. Maybe we caused that issue
- 8:59with
- 9:00uh
- 9:01something that we made that made clients
- 9:04go to the branches. Or maybe that we
- 9:05didn't cause that. Maybe that's a
- 9:06government issue, and the government for
- 9:09some reason required people to go to
- 9:10their branches. We need to find that
- 9:12out. What's causing them to go there?
- 9:15Okay?
- 9:17And or else, of course, there are all
- 9:20those other things that I mentioned,
- 9:21like, for example, we suddenly have less
- 9:25uh space in the branches or something
- 9:27like that. Okay?
- 9:29And how can we improve customer
- 9:30satisfaction given this? Number one,
- 9:34I could solve the problem that drove
- 9:35customers to branches.
- 9:37Um
- 9:38I could increase employees efficiency
- 9:41even if that didn't cause
- 9:43the
- 9:44customers to go to branches. Uh and I
- 9:47could do that
- 9:48by uh training them, or I could do that
- 9:51by getting more efficient uh employees
- 9:54or something like that, okay? And I
- 9:56could automate processes. So, I could
- 9:58either
- 9:59have customers do
- 10:01their pros- their services in ATM
- 10:05machines or something like that, or I
- 10:07could
- 10:08automate uh back office processes and
- 10:11have my employees do them quicker. So,
- 10:13customer sits in front of them, and they
- 10:15only have to press two or three buttons
- 10:17instead of pressing several buttons,
- 10:18calling people, deciding stuff on spot,
- 10:21okay? And next,
- 10:24I could have more services done online.
- 10:27So, people could do those services
- 10:29online rather than going to branches,
- 10:31and I could even incentive in- incentive
- 10:34them to go for online services uh
- 10:38by charging less for that or something
- 10:39like that, okay?
- 10:41Um this is my answer for this. I hope
- 10:44this helped you, and I'll see you in the
- 10:45next one.
About this transcript
This page contains the full transcript of Case Interview Fundamentals: Lesson 33 [Drill: Crowded bank branches] by CraftingCases, generated from the public captions YouTube serves with the video. The transcript has 1,699 words across 292 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.
What you can do with it
Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to choose a caption track available on the video.
Free YouTube transcript tool
YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, caption language selection, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.