This Friday I just finished my Power Day. In the afternoon I got the email saying I didn't need another round, and not long after that the recruiter called that same afternoon to tell me I passed. But since it was Friday, HR usually doesn't get back right away, so they told me they'd first show my results to the HR they're currently working with to see if anyone wants to pick me up, and if not, they'd try passing me to other HRs who might need someone. If there's interest, they'll set up a simple call to get to know each other and go from there. I just hope the matching happens soon.
Timeline:
Early January: phone call with the recruiter — they said last year's candidates still hadn't all been processed, so it would have to wait until February to schedule the mini-case.
Late January: recruiter emailed asking about my availability in February, and we scheduled the mini-case for mid-February.
Mid-February: mini-case in the morning, got a call that afternoon saying I passed, and they mentioned there happened to be a Power Day slot the following Friday — asked if I wanted it. Since I wanted to grab the spot as soon as possible, I said yes right away.
The following week: Power Day in the morning, got the pass notification that afternoon.
(For the mini-case interview report, see my previous post.)
Power Day
First interview:
The first one that day was a product/business interview. The behavioral question was about an accomplishment. Then: recommend my favorite digital app (with reasons), and assume the interviewer knows nothing about it — explain the app and its business model to them (I really did explain even the most basic things).
Asked about the target audience and competitors.
Asked how it makes money (had to name the revenue channels).
Asked what I thought might be a promising future revenue stream, then asked which three metrics I'd want to track for it (besides naming the metrics, I also explained why I picked each one and what the purpose of that metric was).
Asked for six areas I'd want to improve (besides listing the areas, I also explained why I wanted to improve each one), then asked which of the six I'd want to tackle first, and how I'd know if the improvement actually worked (this is where I reused the metric from before). Then asked what I'd say if challenged that the improvement would cost too much money.
After all that there basically wasn't much time left, so I didn't get asked the "10 improvements" question.
Second interview:
The network case that gets shared on the forum a lot.
Asked about revenue and costs. Given data: revenue $40/month, free for the first three months; cost $25/month; installation cost $35/customer; Marketing & Overhead totaling $120/customer; 10K new customers acquired per year. Asked for net value per new customer. At first no contract term was given, so I asked and was told 15 months. -> -$50/new customer.
Then: assume the contract term changes to 18 months — does net value change? At first I said no, and was pushed back on ("are you sure?"). I eventually took a moment and answered that since the base profit is (40-15) = 25/month, a longer term means net value improves.
Now the contract term is 21 months, still free for the first three months, 10% of people break the contract and get charged a $100 penalty fee, and the original Marketing & Overhead now splits into Marketing at $20/customer (variable cost) and Overhead at $100/customer * 10K = $1M (fixed cost). Asked how many customers need to break even. -> 6,667 customers.
It felt like I solved it fast enough that there was still a lot of time left, so they kept going with two graphing questions: first, plot the relationship between demand (customers) and price. Second, plot the relationship between total profit and profit.
Third interview:
The weather insurance case that's been shared a lot recently. Even though people keep saying this one comes up often, I hadn't had time to fully prepare for it beforehand.
Asked who the target customer is and what problem this solves for them.
Shared data on screen: premium prepaid for 12 months at $30/month; servicing cost $3/month; cost of benefits paid if there's a failure: $8,000; regulatory expense $4/quarter + $300 if a benefit is paid. Asked for the maximum likelihood — at this point I asked whether they meant the claim rate needed to break even, and the interviewer was nice enough to say yes, roughly. -> 3.7%.
Shared the risk for A, B, C, D and their cumulative risk, asked which ones to choose to maximize profit, and to explain why -> AB.
Graphing question: given A's profit, asked what the profit would look like after adding B, C, D. This one was simple — I let the interviewer draw it and just told them how it compared to the previous one.
There were still 20 minutes left at the end. They asked if I had any questions, I said no, and that was it.
A few tips on solving cases that I want to share:
After they finish stating the question, briefly summarize what I heard and repeat the numbers back to confirm I understood them correctly, and also restate what the question is actually asking.
When working through a math question, I tell the interviewer to give me a moment to sort out my logic, then briefly walk them through my calculation approach and ask if it makes sense. If it's fine, then I actually crunch the numbers, and afterward I explain the full calculation process and give the answer.
Because I practiced a lot of case reports shared on the forum, sometimes the shared write-ups aren't complete, so you may need to ask questions to clarify or get extra information. My suggestion is to solve based on whatever the interviewer actually gives you in the moment. Sometimes the interviewer doesn't give certain conditions, and the case report you studied might not have covered that either — in that case you need to proactively ask questions to clarify your doubts, and sometimes the interviewer will realize on their own that the question wasn't fully specified and will add a hint that you only need to consider the simple case, etc.
I hope this write-up helps. And I hope things go smoothly with getting picked up and matched!
Discussion
Loading comments…