Scheduling 101: Remember the Basics
Unquestionably scheduling is a challenge in just about any project. It becomes an even greater challenge when requirements change and resources are shared amongst multiple projects and between projects and operational activities.
My first project management job was on the Polaris project. A huge program but one in which most of the resources were dedicated full time and priorities and scope were stable. The hardest part of scheduling in that project was effort estimating and accounting for risk.
Since then I have found myself involved with projects in which resources are shared, there is little or no multi-project resource management and in which clients and users are up to their necks with day to day work so getting them to take part in requirements definition, decision making, testing and roll-out activities involves prayer, threats and the like.
So how do we schedule under those adverse conditions?
Well, one choice is to not schedule at all. Just do the work and finish when you’re finished. In this mode, frustration and escalation as well as diligence and commitment are the drivers for getting the job done. Sometimes projects never get done at all, staying on the back burner forever.
This is not my first choice. Like many who find themselves managing projects I want a bit more control. Clients and executives want to know when the results will be ready.
I have pretty much given up on certainty (uncertainty and impermanence are the only certainties). No amount of the best scheduling is going to provide a guaranteed end date. But, some reasonably accurate sense of “when” is needed and possible. That’s the message we need to get across to stakeholders and ourselves.
The key is realism. Begin with a solid sense of scope and a sense of how stable its definition is. Then create an estimate of how much effort is required. You’ll know this based on having done the work before and by breaking the work down into manageable chunks and estimating the chunks. If the project is unique, chances are many of the tasks are not that unique. Breaking the work down will enable you to identify the uncertainties and do a better job of estimating. If the project is like many others you’ve done before, life is easy. Do a top down analogous estimate. For tasks that have long lead times or are being done by people who are not effectively accountable (for example outside auditors, permit grantors, etc.) Build in a cushion.
As we know from basic training in PM the next step is sequencing to identify which tasks can be done in parallel and which must be done serially.
Next comes the hard part, resource allocation. If you are in an environment that maintains an accurate picture of how resources are allocated then this part is relatively easy. You can see who’s available when and schedule accordingly.
If on the other hand you are among the many without such an advantage you must be careful to estimate resource availability. Don’t fall into the trap of creating a schedule based on wishful thinking (or non-thinking). If you and other resources are bouncing from project to project to operational activities and back, build or if resources like clients, senior managers and end users (yes, these are project human resources) are hard to pin down for meetings and their task durations, then build contingency into your schedule. If you can, minimize interruptions and schedule projects or tasks so that they can be done without unnecessary starts and stops. Don’t be afraid to expand the schedule to accommodate multi-tasking, if you cannot get dedicated resources.
Manage risk. Keep in mind the difference between what is most likely to happen as opposed to what you would most like to have happen.
To account for uncertainty, use buffers or contingency funds rather than elongating individual tasks. At the task level, it is best to pin performers down to a specified target date. On the project level, it is realistic to expect completion within a range. Use contingency funds to manage task slippage and expectations.
Remember, scheduling is a means for managing expectations and performance. Make the initial schedule realistic and aggressive.
Be courageous. Don’t cave in to demands to set a target date that you know is a pipe dream. Adjust the schedule as actual performance and current perspectives dictate.
In the end, scheduling, and planning in general, does not determine the project outcome. The plan sets a direction and approach. They provide the map we use to determine if we are on course. If we are not on course we adjust. If the map is inaccurate we change it.
Don’t forget to leave your comments below.