Decision drift: why your team revisits the same call two sprints later

In May, your team decided to ship workflow automation in Q3 over the integration layer. The leadership team agreed. Engineering agreed. Design flagged a concern but signed off. Customer Success raised the enterprise demand for integrations and was overruled, gracefully, because the competitive timing on workflow felt sharper.

In August, the team is having the same conversation again.

Not because anyone has changed their mind, exactly. Because no one can find the rationale. The Slack thread where the trade-off was discussed has been archived. The PM who made the call has moved teams. The engineering lead who weighed in is on parental leave. What remains is a roadmap that says "ship workflow automation in Q3" and a Customer Success team that's hearing the same enterprise asks they were hearing in May, and now the team is back at the table, asking the same question.

This is decision drift. It's not a disagreement problem. It's an archaeology problem.

The standard response is "document decisions better." Every product team has heard this advice. Every product team has tried it. The Notion doc gets written, gets shared, gets read by three people, and gets buried under the next quarter's docs. The decision is captured, but the capture is in a place no one looks when the question comes back.

The fix isn't more documentation. It's documentation that lives where the decision lives. The trade-off that surfaced in the roadmap review belongs in the calendar event for that roadmap review, not in a folder. The concern Design flagged belongs next to the call that overruled it. The enterprise asks Customer Success raised belong attached to the meeting where the call was made.

When the question comes back in August, the team that captured well doesn't go searching. They open the May calendar event. The rationale is there. The voices that contributed are there. The follow-up question that drew out the real concern is there. The team doesn't relitigate. They update.

The teams that handle this well treat the decision trail as a byproduct of the decision process, not a separate exercise. The async window that gathered input becomes the artifact that captured rationale. The synthesis the meeting produced becomes the record that survives the next reorg. The decision and its archaeology live in the same place.

If your team has had the same conversation twice this year, the test is simple: where would you look to find what the original call was based on? If the answer is "I'd ask someone who was in the room," that's drift waiting to happen. If the answer is "I'd open the calendar event," the decision is durable.