Personal experience interview on initiative, strong and weak

Candidate: During a ten-week internship in the finance team of a logistics company, I was hired to update supplier records. In my second week I noticed that about 3 percent of the 1,400 invoices we processed each month had errors, roughly 42 a month, and each took about 2 hours to fix. No one owned the problem, so I asked my manager for two hours a week to work on it. I sorted a month of errors by cause, found that most came from a few mismatched fields, and built a checking sheet that flagged them before an invoice was paid. By the end of the internship the error rate was down to 0.5 percent, about 7 a month, and the team still uses the sheet.

Structure 5 of 5: Sizes the problem in the first answer, says why it was not the candidate's job, and ends with before and after numbers.

Candidate: I pulled every invoice that had been corrected in the previous month from the weekly export files, which gave me 42 cases, and I tagged each one with its cause by hand. It took me two afternoons. Four causes explained most of them, and the biggest was a purchase order number typed into the wrong field.

Ownership 5 of 5: The candidate's own first steps, how long they took, and what they found.

Candidate: Because I could see the cost every day: two people on the team spent part of each afternoon fixing errors, and it was the part of the job everyone disliked. I also made sure it would not slow my own project, which is why I asked for two hours a week rather than just starting.

Structure 4 of 5: The reason comes first, then how the candidate protected the assigned work. Could name the other option considered.

Candidate: I had no access to the billing system, and IT said a request would take 6 weeks. Instead of waiting, I built the sheet on the weekly export files the team already received, so it needed no new access. The trade-off was that it ran once a week, not live, so I asked the team to run it every Monday before the payment run.

Specificity 5 of 5: A real obstacle, the way around it, and the trade-off that came with it.

Candidate: My manager's first reaction was, "Nice idea, but your project comes first." So I showed the tagged list of 42 errors and the 2 hours each one took, which came to about 84 hours a month, and I promised a short update every Friday. After the first Friday, my manager asked me to present it to the whole team.

Ownership 5 of 5: Quotes the doubt, answers it with the numbers, and shows what changed.

Candidate: The error rate fell from 3 percent to 0.5 percent in the two months after the sheet started, while the number of invoices stayed at about 1,400 a month and nothing else in the process changed. That is about 35 fewer errors a month, or roughly 70 hours of rework saved.

Evidence 5 of 5: Before and after, a check that nothing else changed, and the saving in hours.

Candidate: I would involve the two people who fixed the errors from the first day. I built the first version alone, and in its first week they found two causes I had missed. Now I start by asking the people who do the work what goes wrong.

Reflection 5 of 5: A specific change, what prompted it, and the habit it became.