If the interviewees feel safe, they will tell you their experience and it typically is spot on. If they don’t feel safe, then they may bend the truth a bit. The data doesn’t have to be perfect. The goal here is for the data to be directionally right. It doesn’t matter if the estimates are 1 to 2 hours/days off. It doesn’t have to be precise with this approach. What we find is that the wait time, redundancy, and waste is so much bigger! If there are questions about something, you can pull the data if it’s available, especially data for baseline and post-improvement.