Wednesday, April 29, 2015

The Average Lifespan of a Crisis

Have you ever asked your team to kick it into high-gear for an extended period of time in order to meet some major challenge?  Let me guess what happened…
They rose to the challenge!   Everyone stayed motivated, focused, and maybe even put in some overtime!  Great strides were made towards completing the goals.  But then after a week or so, that high-gear began to slip.  All that motivation and focus dwindled away.  A few weeks later you looked back and wondered if the project was any farther along than it would have been if you had just kept the status-quo.  
How did I know?  Because it always works that way.  "Why," you ask?  Well, it’s not because all developers are lazy.  Allow me explain.

Crisis-mode only works for short periods of time—usually about a week.  Yes, that’s partly because people aren’t capable of maintaining that kind of stress level for a long period of time; eventually it becomes the new normal as the scaryness of the looming deadline wanes.  That's human nature.  But, more than that, the main reason that crisis-mode for extended periods of time never works is because crisis-mode is more about prioritization than it is about being in high-gear.

When you ask someone to drop everything and go into crisis-mode, you are asking them to change their priorities.  You are asking them to focus on the one critical task at the expense of everything else.  The problem, though, is that the longer those other non-critical tasks are ignored, the higher of a priority they will become.  If they are ignored long enough, they too will become crises.

When the priorities necessarily shift back to the "non-critical" tasks, that will leave less time for the critical one.  As your team plays catch-up on their "non-critical" tasks, the critical project will move ahead even slower than it normally would.  Since doing things out of order can make the whole process take longer, artifically forcing everyone to put an ultra-high priority on one task may even inadvertently cause that task to take longer than it would have otherwise.

So what is the solution?  You need to be judicious about invoking crisis-mode.  When you raise the DEFCON level, you should only do so for a predetermined short period of time.  If possible, announce the period of time at the outset so that everyone can make plans around it.  Make sure that the period of time is reasonable--certainly not more than two weeks.  Expect and plan for the inevitable slow-down which will occur in its wake.

What do you do if one or two weeks of crisis-mode is insufficient?  There is no easy answer for that question, but a longer crisis-mode certainly isn't the answer.  It just doesn't work.  You can, of course, fool yourself; you can convince yourself that somehow this time you'll pull it off, but that's just fantasy land.  Fooling yourself will not make you a better manager.  You need to face reality and seek an effective solution.
  • Can the scope of the project be reduced?  
  • Can the deadline be postponed?  
  • Can someone else take over some of the unrelated tasks that are consuming your team's time?  
  • Are there any remaining tasks on the project which could benefit from additional man-power?  
These are the kinds of alternatives you should be considering.  No, they aren't easy solutions.  Any one of those things will look worse for you than a heroic extended crisis-mode effort, but which will look worse for you in the end--success or failure?

No comments:

Post a Comment