We were six weeks out from a committed launch and the engineering lead told me the timeline was impossible — they needed ten more weeks. I immediately set up a meeting with all the stakeholders, communicated the risk transparently, and worked with the team to build a revised plan. I updated the program doc, adjusted the milestones, and kept leadership informed throughout. We launched eight weeks later than originally planned, but the launch was clean with no critical defects. Leadership appreciated the transparency and the team felt supported through a difficult situation.
When the engineering lead told me the timeline was impossible, I did not accept that framing. I pulled the critical path myself, mapped every dependency across four teams, and found that two of the five workstreams had over forty percent buffer baked in. I brought that analysis back to the lead and proposed descoping one non-customer-facing feature entirely. That freed up two engineers for six weeks. We launched three days late instead of ten weeks late. After commit, I ran a retrospective and documented a scoping standard we now use at the start of every program to catch over-estimation early.