The Good Idea Fairy Needs To Be Shot

Wanted, Dead or Alive: Flying fat man in a tutu carrying a light bulb on a stick,

for crimes against intelligence and common sense.

When you turn a kid loose in a candy store, what happens?  They fill their bag with every type of candy within reach.  Kids don't have much of a concept of money.  The have even less of a concept of money in relation to items.  Kids think cars cost $100 because that's the biggest number they can think of at the time.  A bulging sack of candy that costs $40 won't phase them because they don't realize that price is high for junk food.  All that kid is concerned with is getting candy in large amounts.  Likely he or she would pick out a few things that they don't even like along the way, because they're caught up in the moment, but that doesn't matter either, because the sack is bigger with the bad stuff than it is without it.

 

As it happens, there's a very similar occurence when a general learns that he likes analysis.  He orders every type of analysis he can think of, in large quantities, to be delivered as soon as possible, without regard to the resource cost.  We call this an attack of the Good Idea Fairy.

 

Once a general officer's thirst for quality analysis has been stoked, there's little hope of quenching it.  Every thing said general, or his staff, can think to "analyze", they task out to some one.  These strokes of brilliance are often accompanied by the phrase "Hey!  That's a good idea!"  Thus, the phenomenon is named. 

 

The Good Idea Fairy leaves in its wake a path of destruction unparalleled in the annals of history.  Humongous Power Point presentations, exhausted analysts, usually lots of scratched heads.  The time, money, and brain power spent fighting the Good Idea Fairy is uncountable.  Successfully fending off a Good Idea Fairy attack is rewardable with a medal for valor. 

 

The dangers of the Good Idea Fairy are legion.  One: not everything can be, or needs to be, rigorously, analytically, examined.  The Good Idea Fairy does not believe this to be the case, and blesses people with pseudo-epiphanies every day.  For instance:

 

General: "Tell me, good analyst, what will happen if I put this loaded pistol to my head, disengage the safety, and pull the trigger?"

 

Analyst: "Well, sir, you would likely die."

 

General: "Yes, but how likely?"

 

Analyst: "Well, I determine the probability of death, P(Death), to be roughly 1 - P(dud) - P(misfire) - P(miracle).  Further analysis would be required to determine the exact probability, but its safe to say probability of death would be very close to 100%."

 

General: "Hmm.  Go do some research and come back to me with an answer to the twelfth decimal place.  Anything less isn't accurate enough for this critical information.  Have the answer on my desk in an hour."

 

Good Idea Fairy: *cackles gleefully*

 

Often requests to analyze problems that can be solved with common sense are born out of fear or ignorance.  Fear, because last time someone got burned and they're going to make darn certain it doesn't happen again.  Ignorance, because the requesting authority has never done mathematical analysis, and has no concept of it's limitations or proper uses.  Of course the problem with spending resources, i.e. analysts, on problems that do not require rigorous analysis is that it's a waste of time.  Wasting time makes the important initiatives slip to the right (take longer to complete).  Timelines for important intiatives slipping to the right usually isn't accepted in a combat zone, so instead, extra hours must be dedicated towards both projects, meaning long nights and more work for the analysts doing the job, causing burn out, fatigue, crazed rampages through the office, etc.  All because a trivial problem had to be over-analyzed.

 

The key to combating this is communication.  You have to be able to explain to the requesting authority why a simple answer is good enough, and the impacts investigating a complex answer will have to all the other work being done.  Sometimes this works.  Often it doesn't.  The usual solution is to just go with it for the majority of cases, only pushing back when the analysis would be particularly asinine, or the analysts are unusually busy.

 

Danger number two: when the Good Idea Fairy strikes, and analysis requests are flying fast and thick, it is INEVITABLE that at least one, and usually more, of those requests will be impossible to fulfill.

 

An Army officer back home told me a story once about metrics the command in Iraq was trying to use to determine how secure Baghdad was during the middle portion of the war.  One idea was to count the number of kids playing sports (basketball and socccer, mostly) in a public area, like parks or school playgrounds.  The thought was that if kids were outside playing, then their parents must think the area is safe.  More kids outside would indicate an improving security situation. 

 

Now, the premise is fine.  More children playing outside WOULD be a good indication that their parents thought the area to be safe.  But, how, with a force of ~100,000 troops, would it be possible to accurately count children playing outside in a city of 6 million people?  What time of day should they be counted?  What day of the week?  Should they be counted only when it's perfectly sunny outside, or in any weather?  How often does this number need to be reported?  How many people should be diverted from security patrols in order to perform this task?  You can quickly gather how this task would be nearly impossible to accomplish.  Furthermore, the information sought through this method (how secure IS the city?) could likely be gathered in simpler ways, saving lots of wasted effort.  In this instance, other metrics were utilized, and the analyst responsible was revered as a genius amongst soldiers.

 

The key to fixing this problem, in my experience, is demonstration.  This is unfortunate, as demonstration requires you to do all the work in order to show how stupid a solution is before you can get leadership to decide you don't need to do all the work that you just did to get them to decide you didn't need to do all the work to begin with.

 

Still with me?

Demonstration is the only course of action because the word "impossible" does not exist in the lexicon of the U.S. Army.  The closest you can come is "infeasible".  Infeasible is smart speak for "waste of time" to a general, which is exactly the acknowledgment you're going for.  So the quicker you can demonstrate the infeasibility of a suggested course of action, the quicker you can get back to the real work.

 

A personal story.  Just last night I was waylaid by this monster.  Walking back to the office from a marathon meeting, a senior leader recognized his chance to grab three ORSAs and bounce his ideas off us.  He talked excitedly about these ideas he had come up with, analyzed himself on a napkin (literally), and already talked to the commander about.  He was convinced of his idea's feasibility, and demanded that our office set aside time to run the numbers and confirm his stance.  Meanwhile, we're shaking our heads, telling him that there's no way he will be able to accomplish what he's after, that we've already done similar analysis that invalidates everything he's saying, and that further analysis isn't a good use of our time.  But it was too late, the good idea seed had been planted and we were stuck.  Now we've got at least 3 days of modeling, analyzing, prepping and briefing ahead of us, for a problem we already know the answer to.  ARGH!

 

In this case, we tried communication.  Unfortunately, this method of attack did not work, so we move to demonstration.  Though this method will take a lot of work, if properly executed, we can put this problem to bed without being tasked to do anything further.  At this point, that's the best option.  We know the three days required for a proper "demonstration" solution are already sunk, but we'd like to stop the bleeding right there.  

 

So, if anyone sees a little winged person carrying a stick topped by a light bulb, please kill it.  The bounty on this creature's head is high, and you will be rewarded handsomely.  I even think it's possible that with this menace defeated, I may be able to end my tour early.  

All for tonight.  Out here.