Tell me about a time you changed a roadmap because of something you learned late.

0:00So, um, at my last company, Fernway, I was the product manager for onboarding, and we were about three months into building a completely new onboarding flow for new customers. We’d planned it for the whole quarter, and like, design had already finished most of the screens, and engineering had started on the first part.

0:24Then in the second month our support team flagged something odd. They said most of the drop-offs they were hearing about weren’t happening in onboarding at all. They were happening at the first payment, when people added a card and tried to pay for the first time. So we looked at the data together and it was true, about 60% of new users were dropping at the payment step.

0:52I took that to my director and we, um, agreed to pause onboarding and move two engineers onto payments for six weeks.1 It was hard, because you know, the design team had put a lot into it, and some of them felt we were throwing their work away, so there were a few difficult conversations along the way.2

1:20In the end, payment completion went up from 41% to 58% in six weeks,3 which was the biggest jump the team had seen that year. We picked onboarding back up the next quarter, and because the payment step was fixed by then, the new onboarding flow actually performed better than we’d expected when it launched.