Problem Management the Dr. House way

House, M.D is a television show currently broadcast by Fox Networks. The show is named after the main character Dr. Gregory House; portrayed by the actor Hugh Laurie.  Dr. House works as the Chief of Diagnostic Medicine at the fictional Princeton-Plainsboro Teaching Hospital, where he heads a team of diagnosticians.  

I have watched the show for several years now and have gained some valuable insights into IT operations from this show.  While I don't agree with a lot of Dr. House's personality traits, he does have some methods that I think are extremely effective.

 

Don't Schedule Meetings

When the team meets, the meeting is never scheduled nor is there an allotted duration for the meeting.  The meeting happens only when there is a need.  A typical scenario on the show can be summarized by the patient having some sort of medical emergency (an Incident).  For example their heart stops, blood is secreted out of their eyes (gore boosts ratings), or perhaps they lose feelings in their legs.  Once the immediate incident is restored (heart is beating again), the team will then come together and add the new symptom to the board and discuss the impact of the new symptom to the previous hypothesis.  There are several aspects to this diagnostic style that I would like to highlight, but for the purpose of this particular point, I wanted to focus on the meeting itself. 

When you currently schedule a meeting, the common practice is to look at all the attendees calendars, and find a time when they are not having another meeting and are marked as available.  The problem with this method is that it is impossible to determine what someone will be doing at the time of the scheduled meeting.  What if the people in the meeting are right in the middle of some important work.  Those of us in IT know that most of the real work that happens isn't spread out evenly throughout our work day.  There is usually a lot of unproductive time, and then a small percentage of time where work actually occurs.  This time is precious and it is difficult to predict when that will occur.  Interruptions are a major source of waste in IT.

Meetings are important, but scheduling a meeting is a sure way to lower productivity.  A meeting should occur ad-hoc and on-demand and should not be scheduled or have a end time.  Meetings are typically in increments of 30 minutes because our calendaring programs default it that way.  Why can't a meeting be 23 minutes?  or 51 minutes?

The biggest advantage to the House style meeting is that once there is a plan in place to progress, the meeting ends (regardless of time elapsed) and the team goes to work.  They don't continue to discuss other issues or attempt to fill up their 'meeting time'.  They simply dismiss and go work.

 

Visual Tools

House typically writes on a large easel pad of paper during the meetings.  As the case progresses, new symptoms are added here and the diagnosis that are proved false are crossed off.  This visual style of problem solving is very effective and triggers thoughts and ideas that would be difficult to form without the visual stimulus.

 

Step away and do something physical

The solution to the problem usually happens when you are away from the problem.  Throughout the episode, House is absorbing all of the data being gathered to solve the case.  He is usually never standing still and thinking.  You can typically find him bouncing a ball against a wall, playing music, or some other form of physical activity.  This may be viewed as House being distracted, but it usually results in an aha moment where a key connection is made and the answer to the puzzle is revealed.  I don't know much about the science of brains, but I have found from personal experience that when I am trying to solve a complex problem, sitting there staring at the problem and trying to think about directly doesn't always work.  Walking away and doing something physical tends to trigger the right things that allow my brain to come up with the solution.  I've solved lots of difficult problems while in the shower.

 

Challenge the obvious

Problems are identified as problems because they are not obvious.  The cases that House selects are selected because nobody else can figure them out.  The obvious solution is usually not the root cause.  This may seem contradictory to common sense, or to the Occam's Razor principle, but in the case of Problem Management, the root cause is usually obscured and will not be easy to find.

 

Test quickly and often

As soon as anyone on the team has a reasonable hypothesis, it is immediately tested.  Either to rule it out, or to confirm that they are on the right path.  The test adds data to the overall diagnosis and usually results in additional symptoms being discovered or introduced.  Even if the test is unsuccessful for proving the original hypothesis, it always helps to uncover the actual root cause.  Don't wait around for more occurrences, come up with a reasonable guess now, and then test to confirm.

 

Trust but Verify

One of the controversial personality traits of Dr. House is that he assumes everyone is lying.  A common investigative technique for his team is to go to the patient's home without their knowledge and break into their home.  The team is looking for environmental clues that will explain the symptoms.

I won't go as far as saying your customers are lying to you, but I will admit that there are times when the whole truth isn't divulged.  This can be intentional or simply as a matter of ignorance.  Whatever the case, you should witness the behavior yourself.  In the case of a problem that you are trying to solve, you should see it firsthand.  Go watch a user recreate the problem.  Watch it with your own eyes.  Often times you will see something that was lost in the translation by the time it got to you.

House's bedside manner is notoriously crass.  He tends to avoid meeting the actual patient until the very end.  Part of this is because he cares more about the science and the diagnosing the case than the patient, the other reason is because he likes to observe.  So while he doesn't usually interact directly with the patient, he does like to watch from a distance.  He is often watching from outside the room or from an observatory deck above the operating room.

 

Rule out what is NOT the problem first

House will often have his team perform a procedure or administer a dose of drugs that will eliminate one of the assumptions made.  The intention is to rule out things in order to narrow the list of potential causes.  Throughout the process of elimination, new symptoms are raised and added to the overall evidence for the diagnosis.

 

If you have ever watched the show, I would love to hear your feedback on my analysis.  Watch it again with these ideas in mind to see if I am right.