Imagine the situation. You run a pizza store and you are losing money. Your raw material costs are far more than they should be. You jump to the most obvious conclusion:
Somebody must be stealing the cheese. (Either that or you have king sized mice wondering about).
After spending a lot of money on an expensive cheese alarm and CCTV system it becomes abundantly clear that nobody is smuggling cheese out of the back door. And you didn’t spot any big mice either.
There is an analytical solution to the problem. It is called M.E.C.E. (Mutually Exclusive, Collectively Exhaustive).
Mutually Exclusive
First you need to back up to the point you are totally sure you can define the problem; in this case raw material costs are more than they should be (not somebody is stealing the cheese). Then create a list of the things that could be causing the problem. The raw materials that you are spending too much on:
- Dough
- Tomato Sauce
- Cheese
- Topping
- Packaging
The categories should be “mutually exclusive” to avoid double counting and confusion. Topping could include tomato sauce and cheese, it depends which school of pizza making you come from. “Other Toppings” would be a far better description. Nobody will get confused.
Now you have defined the categories re-evaluate the problem. Where are you spending too much money? If the packaging, other toppings, cheese and tomato sauce are OK then it must be the dough.
Completely Exhaustive
Drill down and repeat the exercise; list all the reasons why you could be loosing money on dough.
- You aren’t receiving all the materials you paid for
- The mix is wrong, too much expensive flour, not enough cheap water
- You are throwing it away, too much in the bin
- You are giving it away, thick crust not thin
- Somebody is stealing it
- Other
Other is the “Completely Exhaustive” bit of M.E.C.E. We don’t and can’t know everything. There is always an “other” category. It is the balancing category, the bit that is left when you have counted everything else. Realising that there is always an “other” means that you won’t ever miss anything.
- If other is small don’t worry about it
- If other is big then you need to dig into it
And Repeat
Keep creating “mutually exclusive, completely exhaustive” categories and drilling down until you can pin point the problem, and once you have you can do something about it.
Alternatively you can of course keep jumping to conclusions and install a bunch of very big, very expensive mousetraps.
Read another opinion
Image by j/f/photos
Hi James,
Thinking and investigation versus money and ‘mousetraps’ wins every time.
The old saying ‘An ounce of prevention is worth a pound of cure’ sums up what you are saying nicely.
Adrian
Hello James
I believe that McKinsey made this one famous – at least one of their ex-consultants did. It is great as it goes and it does have its limitations. Being a physicist I would say that it is Newtonian paradigm rather than the Quantum Physics paradigm. The MECE formula sees the world through reductionism (parts) and thus is blind to the ‘systems perspective’ that sees the interaction of parts to generate emergent phenomena. So it is quite possible that you will arrive at a faulty hypothesis because the MECE does not allow the relationship/interaction of the parts to generate the outcome.
The next point I’d like to add is that of validating (actually invalidating) any hypothesis we generate. Karl pointed out that you if all you have seen for all of time is white swans then you cannot conclude that there are only white swans. The possibility exists, no matter, how small that there is a black swan out there. Now add to that the human bias (Tversky, Kahneman) to look for what confirms their prejudices, pet theories or frameworks and you end up with the following:
Whatevery hypothesis you come up with for the root cause set-out to find evidence that disproves it! If and only if you cannot find that evidence after diligent searching then you can conclude that it is highly likely that you have arrived at the right cause.
Now the thing is that if you came up with the hypothesis then you are not the best person to go around looking for evidence to disprove it. So it helps if that task can be fulfilled by someone who disagrees with your hypothess.
Maz
Thanks for your comment Maz, I can’t and won’t argue with what you say, I suspect it would be foolish of me to try.
I can’t help but think though that it is better to be approximately right than precisely wrong. Outside of the world of mathematics there are very few categoric proofs
There is a fascinating post here, which demonstrates quite how far off the analytical mark people can get. http://thecustomerblog.co.uk/2012/02/09/three-reasons-why-converting-data-into-valuable-business-results-is-no-picnic/