Software project managers should allow for experimentation

Software projects are creative processes that involve trial and, yes, error, argues PM expert Bas de Baar. However, these errors should be opportunities for learning, not project killers.

Bas de Baar, PM expert
Bas de Baar

What is killing thousands of projects per week? The fact that we have no clue what is happening in our project and its environment, and still act as if we have paranormal gifts that allow us to predict the future. Yeah, I know, I've said this before. We are too stupid to understand reality. So, there I said it. Again.

But what can we do? If we cannot create a Gantt chart, what is there left?

Create your bloody chart. It's not that piece of paper that is the problem. It is you for thinking reality will simply follow it. It's just a communication tool, for crying out loud. Just a piece of reflection. Not some holy relic.

It's OK to admit you are also human. Don't be ashamed if you don't know how certain tasks will end up. Just try things. Experiment. Wing it. Trial and error. Just do it. And look what happens. Does it bring you in the right direction? If so, do more of it. If not, try something else.

It's OK to admit you are also human. Don't be ashamed if you don't know how certain tasks will end up.
,

The solution is so simple. But somehow, "experimenting" lost its appeal. Management (with a capital "M") doesn't allow uncertainties. The whole quality hype didn't help us either, with its "zero defects" and "doing it right the first time" crap. It just means we should think before we do something. It doesn't mean we should kill everyone that needs a second attempt! Heck, even Frederick Brooks wrote in his classic The Mythical Man Month "the management question, therefore, is not whether to build a pilot system and throw it away. You will do that." His advice? Plan to throw one away.

No, this doesn't mean just shooting from the hip. This means careful consideration of which experiments to perform and closely monitor the results with all metrics you have at your project management disposal. This is just normal software PM. Prototypes, proof of concepts, every iteration is a new experiment that brings you either closer to the end result or helps you adjust your path towards the desired result.

"But a builder doesn't need three attempts to build your house!" I hate those kinds of discussions. My primary reaction is, "Go ask your building constructor to build your software!"

Software project management
Estimating actual project progress 

Using iterations to help balance priority and risk 

Project managers can't please everyone 

The Software Project Manager's Bridge to Agility: Scope Management

Of course, if I leave my bias aside, the question from an IT outsider is legit. It is, however, difficult to admit that we as IT professionals are still trying to figure out how this all works; our profession isn't a mature one. In the field of house construction, the professionals are thousand years ahead of us. The ancient Egyptians, Romans and Greeks were all building constructions while the concept of a database wasn't invented yet. I am embarrassed about how ICT is trying to mature its profession; it will largely be a trial-and-error, incremental approach. But, before constructors could build houses that stood erect, pyramids, cathedrals and ordinary houses collapsed over the centuries. Attempts were made. Buildings went down. Trial and error.

So, everyone has done it before. We have to do it, there are no other options. We are already doing it. But we cannot call it "experimenting." We call it "iterative," "evaluating alternatives," "generating options" and so forth.

Could you do me a favor? On your next planning put "First Attempt," "Second Attempt" and so on. End every phase with "Looking for clues." Just try it.

----------------------------------------
About the author: Bas de Baar knows all about the wacky world of project management. He is a project manager in the publishing industry and is editor of a popular Web site devoted to project management, www.SoftwareProjects.org. His venerated instructional book on sudden project management, Surprise! Now You're a Software Project Manager was published in September 2006 and is based on real-life experience.


This was first published in July 2008

Dig deeper on Software Project Management Process

Pro+

Features

Enjoy the benefits of Pro+ membership, learn more and join.

0 comments

Oldest 

Forgot Password?

No problem! Submit your e-mail address below. We'll send you an email containing your password.

Your password has been sent to:

-ADS BY GOOGLE

SearchSOA

TheServerSide

SearchCloudApplications

SearchAWS

SearchBusinessAnalytics

SearchFinancialApplications

SearchHealthIT

Close