The third movie [he worked on] was 2000’s X-Men, which director Bryan Singer was making for Fox. To better understand the genre, [Kevin] Feige [future president of Marvel Studios] immersed himself in Marvel comics. “I did a much deeper dive than I ever had before,” he says. What he discovered was extraordinarily rich. … As Feige consumed stacks of Marvel comics, he wondered why others working on X-Men didn’t do the same. “I would hear people, other executives struggling over a character point, or struggling over how to make a connection, or struggling over how to give even surface-level depth to an action scene or to a character,” Feige recalls. “I’d be sitting there reading the comics going, ‘Look at this. Just do this. This is incredible.” Bloomberg Businessweek, The Master of the Marvel Universe, Apr 7, 2014.

I was just a project manager yet I knew more about running the VP’s department than they did. Why? Because I took the time to dig in and understand how their organization worked and how it had performed in the recent past. When it came to organizing and committing the department to a project it was obvious what they could do and not do.

I recall telling another manager, who was committing a team to my project, that he was not planning for enough time for a major task. The team manager was floored. Since when did anyone say anything other than “you have to do it faster”? I understood his organization. It consistently delivered late and buggy software because the schedules they agreed to were too short. Once we settled on a schedule that matched their past performance, the team did a great job.

How did I get so smart? I simply did a deep dive into how the organization was performing. I looked at every last project I could get my hands on, and there were hundreds of them (the company usually had 20 to 30 projects going on at any one time). I am sure that I’m the only one who ever took the time to look at all the past projects’ archived data. The good news was that the company had a standard for archiving past project data. It was not perfect, but it was good enough.

By the time I had been part of several projects in the corporation, I knew very well how we had performed in the past and what we did well and not so well. As we planned and executed new projects, it was obvious what we could do and what we wanted to avoid.

The problem was that it was not obvious to many others. Instead they would repeat what they had done in the past, both the successful and unsuccessful. My job, it felt like, was to help pick and choose — to help prune — those decisions. If I could get into and influence the key decisions, such as the overall length of the project, I could set the teams up for success, rather than the less than success we’ve had in the past.

I recall presenting a risk analysis that showed, based upon past project performance, that our project had a high risk of missing it’s dates and having too many open defects. The business manager said to me that I must never present such a status again in the future! Instead, he stressed, I must show a risk analysis that says we can make the end date. This way people won’t be confused, he said.

We ended up slipping the project, once — due to a predictable crisis — and aligning the end date with the past performance of the teams as our response to the crisis. We made that date without any further slips. The business manager exclaimed “Bruce, this stuff really works!”

See more on this in Have A Crisis? Hurry, But Do Nothing

If we’ve been successful in the past, but find we are losing our edge, then we should also consider doing a deep dive to get reacquainted with how things really work.

Zuckerberg will need a new hack for that. That’s one of the reasons he’s back to coding every day. Colleagues say he wants to immerse himself in the daily lives of his underlings. If he’s going to keep inventing new ways to keep users coming back … he needs to keep Facebook’s core talent motivated. Bloomberg Businessweek, How Zuck Hacked The Valley, May 27th, 2012.

Consider deep diving into how our organization and project teams perform. This gives us the ability to make key decisions, daily decisions, that drive our projects to success. Otherwise, the decisions we make are often just random noise, that don’t result in projects being anything but the normal, mediocre, late and buggy things we’ve always done in the past.

See related How To Give Your Boss An Accurate Quick Estimate

What have you done lately to deep dive into your team or organization’s past performance?