Tell me about a time you were trying to understand a problem on your team and you had to go down several layers to figure it out. Who did you talk with and what information proved most valuable? How did you use that information to help solve the problem?

I inherited a team as part of my promotion, and when I was getting to know my team and what others thought of my team, i uncovered that one of the team members had some really questionable and often bad feedback. This was perplexing since I had dissatisfied business partners and dissatisfied engineers, and I had known of him to be really hardworking and going above and beyond. I did a few rounds of investigation

1. I first discussed with the project managers, and the feedback I got was that he is never planned since he didnt fully understand what the customer really wants, and so his stories lacked the detail. 
2. I talked to my business partners and they told me, half of the time he doesnt know how to elicit requirements, since he doesnt fully understand their problems. 
3. I finally discussed with my engineering team, and broke my questioning into the types of projects, turns out there was one category of projects he was brilliant at, the ones which were purely technical like ensuring system availability, ensuring stable code, anytime there is an infrastructure change planned, or even massive redesign of systems. 
4. I finally had a discussion about his projects with him, and I asked what is it about these Technical projects you like so much, he first went defensive since he was unsure about how to react to a new boss. But I assured him I was looking out for him, and then he opened up and told me that I used to be a mainframe developer so  he is very comfortable with any project that has a technical bent. but he isnt too comfortable coding anymore.. the size / scale of projects dont daunt him, a non technical savvy audience does.

I realized this is a skill mismatch, and at that time I had a plethora of technical projects, and I assigned him purely on those type of projects, and minimized his business interaction to on a need basis. He was still a valuable talent, and these projects still needed to be done, he was the best fit for those projects. 3 quarters later he was awarded the employee of the quarter for his efforts. 

How I'd use a personal anecdote to tackle this behavioral interview question.

There was this one time when i was managing the Checkout Product at PayU. This product was integrated and used by millions of merchants to accept payments from their customers. The conversions for the checkout page or success rate as we call it in the payments industry was significantly low at 60 % which is lower than the market standard. 

I started doing the analysis on why there were many cancellations or drop offs or even failures in some cases.  To start with, I asked my analytics to collect and provide me with some analytics on PGs configured for the merchants who were using the Checkout. It was possible that a PG not performing could impact the success rate. Another reason could be around missing payment options or customers could not discover the payment options of their choice. So I asked the dev team to add a dialog to get the reason why consumers were cancelling the transactions. Moreover we knew there were some problems such as high page loading time and technical issues so I asked my analytics team and dev teams to add the necessary clevertap events to get these numbers and success rate for cases where these issues were encountered. We also asked them to collect browser and OS level data. We let the numbers get collected over a week’s time and then decide to have a meeting. 

I spoke to the analytics team in detail about the problems. They came up with several reasons for the checkout page not performing well : 1) AMEX PG not performing well and a lot of transactions getting dropped off or cancelled, Some other PGs were doing particularly well for certain payment modes. 2) Page loading time was high on slow networks and that let to customers dropping off or cancelling the payment 3) from the dialog box we added, we found that a lot of customers were not able to find the relevant payment options 4) in some cases, customers had chosen the option which said “ the amount shown on merchant page and our page was different so they cancelled”. Basically we would show the final amount after adding convenience fee. 

Based on this information, I asked the dev team and onboarding teams to focus on quick wins. We configured the PGs that were performing the best for each payment mode and decided to monitor it properly. I also started planning with the architect about building a routing mechanism which would automatically route txns to different PGs based on current success rates. In addition to that, I prioritized adding a payment breakup to the checkout page which was low effort so that the convenience fee was clearly visible so that customers would not cancel thinking the amount on the payment page was wrong. 

Other items required more brainstorming and technical effort which we decide to plan for the succeeding weeks such as feature of showing last used and most used payment mode at the top, showing trending payments on top. Eventually on the back of these changes, in a month’s time, our success rates improved to 65 % and within 6 months, our success rates were touching 75 % which were better than market standard.

@Bijan please evaluate.