4 replies
by dc_guy
The key to this question is: Every time a person has to switch from one activity to another, they lose a little time that can never be recovered.
Obviously there are no formulas for this. There is the process of merely shuffling papers around or logging off and onto different applications; closing off a train of thought and picking up a different one; getting into a different rhythm -- using different skills, memories, attitudes, etc.
All of this is affected by how many different assignments there are to juggle in their head, how long the person gets to be immersed in one before being pulled out, how difficult each one is, and how accurate the work must be.
Each of us reacts differently to theses demands. There is a tiny fraction of the population who can switch tasks almost effortlessly. But for most people it's safe to say that every time they switch from one task to another, they effectively lose at least half an hour.
That's a lot of time. If a person has only two assignments and only works on each one twice during a work day, they lose TWO FULL HOURS: one fourth of the day. You don't see this because it looks like they're working. But they're not working at full speed, they're making little errors that they have to go back and fix (or worse, errors they don't notice), and they're trying to remember that great idea they had yesterday but didn't write down.
You'll get the best work out of people by minimizing the number of assignments. Production support, in particular, is a terrible time destroyer because it pops up in many little pieces scattered throughout the day. A person trying to support an annoying production system at the same time they're trying to build something challenging can end up with effectively two hours of productive work in a day. Or less.
by jschiel
by barnumthomas
The optimum, of course, is to commit individuals to projects from start-to-finish. The reality, however, is that most of us are under-staffed to a point where this cannot be accomplished.
I would suggest to you a rotation system that works in one of two ways:
-Two days on project; three days on operations
-9-12 on operations; 1-4 on projects; 4-5 on operations
The latter method will allow you to have the beginning and end of days for operations while providing a space of three solid hours during the - generally - slower times of the day.
Just some suggestions for thought,
Thomas Carpenter
Senior Consultant - BarnumThomas
http://www.BarnumThomas.com
by hoghat
We are currently on split-team system where one team does support/operations/maintenance and the other development. Our rotation time frame is about 6 months, depending on when projects come to an end. Most of our project due dates are timed with the end of a quarter. Programmers rotate into operations after the completion of a project, basically swapping out positions. This is a good system, it allows us to focus better both on development and support. One drawback I have seen so far is that when staffing the operations team they only measured what it would take to handle support and operations. You need to add in at least one resource dedicated to support elimination/maintenance for re-occuring support nuisances... somebody to fix the holes so you're not just bailing water out of the ship every day. Also, be sure to clearly define expectations of the teams so there is no confusion over who needs to fix something that was installed on the development side for a project that is no longer active.
Where I work the operations team also has it's own manager, so they can focus on managing support or development more clearly as well. All the teams in our business area contribute a portion of their programmers to this operational team. So there are several teams under one roof all focused on support, mostly on their own individual areas but they can share resources if necessary. The manager of that team is also a rotational position. The idea is that the managers gain more exposure to the business area/people outside of their own team.
One thing that doesn't work, putting people on support team and leaving them there. We tried a permanent operations team years back and it was not pretty. Even if you hire new people into the team on support, they will eventually expect to move to development.
This discussion is closed to new replies.
Trending IT Employment posts
Sort by Trending