So I was having lunch with this guy from Mainframe (www.mainframe.ca) and we were talking about stuff: What Creo was like in the good old days, what's up with the local high-tech scene, and of course, the topic of project management, especially in the field of software development. I of course had a bit to say about managing software development projects but I held back for the most part so as to not suck all the oxygen out of the room.

Therefore, it's wonderful to have a blog so one can vent out all the wonderful musings in your head, for I can safely assume dear reader, that you want to be here. Or that you are so stupefied with boredom that you cannot even expend the energy to click your mouse and hyperlinks over to some other web page. Same difference for me.

Anyways, the secret is very simple: Change your 85's to 95's and you are a good project manager. Of course, nobody realizes how important that is and you won't get much credit for it. But trust me, that's what a good project manager does.

Say what? You want elaboration? Oh, okay. I love to talk anyways....

Most people underestimate the fragility of non-redundant complex projects. That is to say, uncertain events that must be accomplished in a sequence are prone to failure to a greater extent than most people intuitively realize. For example, suppose you have a project with three separate main tasks and each task is 85% certain to be achieved in the time allotted. What are the chances of the project being successfully completed on time? Eighty-five percent? Eighty percent? How about seventy percent? WRONG baby, the chances of that project completing on time is 61 % or (85% x 85% x 85%). And in software development, a project with three main tasks is a simple, simple trifle that's not even worthy of a full-time project manager.

Even though the math says that less than 2 out of every 3 projects will be successful with those kind of odds, most people intuitively think the chance of success will be higher. But don't take my word for it, pose some hypothetical questions around your office: Say you have this project with three things that are 80% plus sure of happening. How often should the project manager deliver the goods, and how often should he be allowed to fail? EVERYBODY will say a 75% or higher success rate for the project is a reasonable expectation.

(Allow me an aside here: If you try this at your local employer & you get the right answers from more than one person, please e-mail me IMMEDIATELY so I can send my resume to wherever you work)

The remorseless logic of this simple mathematical equation absolutely means that those who survive in project management are those who focus on the 85's and change them to 95s. That is to say, up the success rate of a project from 61% to 86% (95% x 95% x 95%). This means that for every five projects, four of those will be completed and/or be finished on time.

Now that you know the really big secret of being a good project manager, you are probably thinking hey, this is great, now I can make the leap to becoming a manager instead of being a coder/engineer/starbucks barista. And if you are thinking that, that means you have not yet considered the implications of focusing on just that 10%: Why it sucks.

Here are five reasons:

1. The 85/95 rule means that you don't have to watch out so much for the stupid & lazy people, but the clever gamblers. The really bright people on your team love to do bleeding-edge type development because it's cool and interesting. They don't really think (or care) that moving the success percentage from 85% to 95% of their assigned task is such a big deal. This means that you, as project manager, have to yank them away from the bleeding edge or try to compensate for their gambling elsewhere in the project.
2. If you are doing your job properly, you are only focusing on the critical 10% of 3 or 4 task where you can make a difference. It means that you are not doing any real fun thinking stuff, but mostly carrying water for other people.... Who don't understand why you are kicking up such a fuss over a lousy ten percent.
3. Unless the person above you is a very smart/experienced executive, they are going to get upset with you for expending a LOT of resources & time on the critical 10% of three or four subtasks.
4. At review time, members of your team will think that you contributed "on the margins". That to say, you made an impact on 10% of whatever that subgroup accomplished. So maybe they will give 10% of the credit. So at review you will have 3 or 4 subgroups saying you deserved marginal credit.
5. The really fun projects with big paybacks usually have one task that's at 70% (or lower). Watch what happens when you plug that percentage into the sequence: 70% x 85% x 85% = 51%. So it's a fifty-fifty chance of success. Get used to failure if you want to keep signing up for the fun stuff. More importantly, get your BOSS used to you failing if you want to keep having a job.

In conclusion, there is usually a shortage off project managers in product development groups, especially software. And the math above combined with people's intuitive expectations of success is pretty guaranteed to make sure that shortage extends into the future.

But darn, it sure is cool when a project comes together against the odds and takes on a life of it's own. Even if very few people think you did anything to help make it a success.

Cheers, DJ