“When you are in the middle of a story it isn’t a story at all, but only a confusion; a dark roaring, a blindness, a wreckage of shattered glass and splintered wood; like a house in a whirlwind, or else a boat crushed by the icebergs or swept over the rapids, and all aboard powerless to stop it. It is only afterwards that it becomes anything like a story at all. When you are telling it, to yourself or to someone else.” – from Alias Grace by Margaret Atwood
Whatever project management approach a team uses, sometimes everything falls apart, commonly due to work piling up at the end, but sometimes due to a key individual leaving, or a pivotal assumption no longer holding true, or many other reasons. When that happens, the project can become like a whack-a-mole game, with leads working from issue to issue as they pop up faster and faster.
I served as one of many workstream PMs on one very large project where this didn’t happen. Out of the seeming chaos the multi-million dollar IT project came in on time. Here’s my view of what we did to succeed:
- Had very clear interim milestones that were generally known and served as reference points for discussion. I’d characterize them as a milestone per workstream per month.
- Held seemingly interminable weekly risk/issue discussions. These were open, no holds barred reviews of anything at all that could endanger achieving milestones. Often risk/issue discussions are polite exercises in avoiding the fact that the emperor has no clothes, with team members carefully avoiding forbidden topics. On this project everything was open for discussion.
- The program manager excelled at visibly not sweating the small stuff, directing workstream leadership to handle their localized risks and issues themselves, and focusing program energy only on those he judged to have overall impact.
- Each of the many workstreams had its own project schedule; the program insisted on detail where it mattered but not where the detail was irrelevant. For example, there was a minutely detailed cutover plan for production migration.
Many of us were surprised when the chaos all came together on time with surprisingly few glitches. I attribute this program’s success to the program manager’s unflappable focus on milestones, encouragement of unfettered group risk/issue analysis, and ability to parse program from project concerns.