Building a Project Schedule the Best Case/Worst Case Way [The Liquid Approach: Part 2/6]

Liz Pearce | February 29, 2012

best case worst case
left Image

Building a project schedule should be easy, right? Just lay out the tasks, assign them, figure out how long they’ll take, and boom—you should have a completion date calculated in no time. But would you bet your salary on it?

The truth is that figuring out how long things will take is at best an educated guess. And adding up a bunch of guesses produces a pretty risky timeline—one most people are wary of, so they slap a big buffer on at the end of the schedule and call it a day. Feeling confident yet? If not, have no fear. There’s an easier way to get a more accurate project schedule!

In LiquidPlanner, you provide simple best and worst case estimates of the effort it will take to complete a task. (For example, the task “write requirements document” will take 4 – 6 hours.) Those two numbers, representing 80% confidence levels, work out to a little statement of probability. Put in a whole bunch of ranged estimates and pretty soon you have a statistical picture of when you’re likely to complete the set of tasks.

Don’t worry—you don’t need to have passed college statistics to get a statistically accurate project schedule. Just follow these simple steps and watch your schedule appear before your eyes.

  1. Add your task and assign it to the person who will do the work.
  2. Give the task a best case / worst case estimate of the number of hours of work the task requires.
  3. Prioritize the task along with the other work assigned to that person.
  4. Hover over the schedule bar to see when you’re likely to complete the task.

You’re probably thinking, “That’s all well and good, but what about my deadlines?” There’s a place for deadlines in LP. We call them “promise dates.” You can enter them for each task or project independent of the schedule dates. That way you can let the schedule tell you if the deadline is realistic. This works well in a living system, because today your schedule may be perfectly on track. If tomorrow something changes to put your end date at risk, the system will let you know.

If your promise date is in the red, you can take a number of steps to fix your project schedule:

  1. Add additional resources or reprioritize work.
  2. Make sure that the tasks driving the date out are estimated correctly (i.e., has everything been updated?)
  3. Cut the scope of work.
  4. Re-set expectations with the stakeholder that the date is not realistic.

Uncertainty is an inherent part of projects for all knowledge worker project teams. By capturing the uncertainty associated with task effort, a team can have a more accurate sense of the overall project’s uncertainty than most traditional methods (like project buffers) provide. Furthermore, by exposing that uncertainty early on in the process, steps can be taken to mitigate risks before they impact the project schedule.

Read the previous article in the series: Why Old-School Project Management Is Becoming Extinct
Next in the series: Stop Thinking so Hard: Multi-Project Scheduling Is Easy!