Managing a project vs Managing a Development

Question

Our organisation uses PRINCE2 for every project we run; IT is large part of our organisation but we're not a software house. Whilst I appreciate PRINCE2 is a great method of running projects, and, we've had a lot of success with it, I feel it's fairly high level for the nitty-gritty of building software.

Is it unheard of to run an AGILE method such as SCRUM for the day to day building and deployment of code underneath, or, within the framework of PRINCE2?

Obviously we have a number of qualified, experienced PRINCE2 practitioners in the business so I'd never suggest replacing PRINCE2 with SCRUM, but I can still see the benefits of SCRUM.

Anyone any thoughts, in practice do the two contradict or complicate one another, does anyone have any personal success stories. What about other methodologies? Can we let PRINCE2 take care of the high-level stuff and then run another method for the nuts and bolts of developing?

Thanks

Andrew

Answer 1

One of the biggest parts of a project manager's job is communicating between the stake holders (e.g., upper management) and the project resources (the programmers, etc.).

The risk you run into by using two more more very different Big-M Methodologies is you will have to "retrain" management. They will have different expectations of what you will be delivering to them as interim reports, etc. Especially where one is used almost everywhere else in the organisation. You may very well start looking sloppy to them.

AGILE and SCRUM don't necessarily reduce the planning and analysis that has to go into an IT project, it distributes it throughout the project, and make the programmers feel like they aren't having to do all this yucky planning before they actually build something.

Answer 2

If that's the case, if the project manager speaks both methodologies and can communicate up the chain in PRINCE2, but down the chain with SCRUM, is that workable?

Answer 3

I'd be concerned that the translation from SCRUM to PRINCE2 is imperfect, doubles your work, or is just a PITA.

The only way to know for sure whether this is something you can handle is to test it out. Get clearance from the stakeholders to run a test project like that and see what happens. Compare it to the results of something similarly sized that you've done before. Just make sure it is big enough to have a two or three sprints in it.

Answer 4

@Codeslave - common sense prevails again I suppose :) Thanks for the feedback