2 de junio de 2013

The tale of the tailor, the princess and the continuous improvement plan




Most organizations use mangement by objectives (MBO), key performance indicators (KPI), balance scorecards, continuous improvement programs, etc. All these tools are useful provided they are founded by good paradigms (i.e. based on principles). Every company has to find its balance between P (Production) and PC (Production Capacity). Production capacity includes research, learning, innovation, process improvement, dealing with customers, etc. Production has to do with sales, billing, value creation for investors, etc. If metrics are not based on principles, we could be very efficient measuring and improving everything, but... what if we are climbing the ladder of success only to discover it's leaning against the wrong wall?

What follows is an illustrative text on how metrics and continous improvement plans may lead to a false feeling of good performance. This is an excerpt of the book Slack. Getting past burnout, busywork, and the myth of total efficiency, by Tom DeMarco. It starts with a tailor who lost a needle in a haystack. For some reason, it was important to him to find that needle and worked hard into it, but he was not succeeding at all. This tailor had read many books and tales. When he was up to giving it up, a young princess showed up and then he remembered: A real princess would have the sensitivity to feel a pea through 20 mattresses. She should help him in finding out his needle, for sure! What started as a productive collaboration towards a simple and definite goal, ended up with certain changes along the way. Both of them planned a new of working to be more efficient under a continous improvement plan. However, this kind of plans sometimes tend to maximize the wrong goal and result counterproductive to organizations (and people).

[...]


There once was a young tailor who chanced to lose a needle in a haystack. He sercherd and searched for it, but to no avail. Being of a philosophical bent, he considered, as he searched, the abstract of the matter. He looked to see if this might be a single instance of a larger class of problem. Sure enough, he realized almost at once that the heart of hs present dilemmalay in the superfluity of an obscuring element: The population of hay is large compared to the population of needles. Of course, there could be a boundless set of such problems, all of them at least abstractly identical to his own.


Let n, he thought, represent the number of needles, where n is equal to one in this case. And let h represent the number of discrete pieces of hay. Now, for h larger than n, the difficulty of finding the n can be arbitrarily high. Looked at another way, the probability, P, of finding n goes down as h increases. In fact, there is no number, delta (no matter how small), that is not greater than P for sufficiently large h. In proceeding through this logic, the young man was discovering what our later era would call the Goldbach Conjecture, essential to the science of fractals and Mandelbrot sets. But of this, he was unaware. He was a tailor. While he had solved the relationship between P, h, n, and delta, he was no closer to solving the underlying problem, since he still couldn't find the needle.

As luck would have it, there happened to be knoking about that very haystack at that very moment a lovely young princess. Normally, the tailor would have minded his own business, not even dared to raise his eyes to look at her. But hen he had an inspiration:

“Say,” he said, “I see that you are a beautiful princess, and I know a thing or two about beautiful princesses.”
“Yes,” she sighed, “it's an occupational hazard. You aren't in the princess line for more than a few moments before everyone is thinking they know all about you.”
“I know, for example, that a princess can feel, even through as many as seven mattresses, that someone has peed in the bed.”
“Mm,” she said. “You've got some of the details wrong there, but I guess you've caught the essence: We princesses are a sensitive lot.”


Then he explained to her about the problem of the needle in the haystack.

“I think I see what you're getting at,” she said. “You want me to lie down on this haystack and help you locate your needle.”
“Right”
“I'll do it. We should have your needle in a trice. Now where did you say you lost it?”
“Of that I am not too sure. But somewhere in this general area.” And he waved his hands over an expanse of haystack encompassing very large h.
“Mm. Well, let's give it a try.” With that she plopped down on a likely spot, wiggled slightly into the hay, and let her eyes drift. “Nope. Not here. Let's give a try over this way.”

She moved to a different part of the haystack and lay down again. Again her eyes wandered dreamily. The tailor felt a thumping in his chest. He realized he had never known anything about princesses before, at least not about this princess, not a thing. And now... Now his life was transformed; it could never be the same again. Where before there was mending and darning, now there was the possibility of Love. His days might be filled with beauty and enchantment, and dancing and cuddling. That was the good part. There was a not-so-good part as well: Where before he had had nothing to lose but a dumb needle, now he had everything to lose. He felt a sudden panic. Oh, above all, don't blow it, he thought.

The princess was frowning. His princess was frowning. He felt his rosy new future slipping away. “Well, it's not here eigher,” she said. “I can't understand it.”
“No matter, really,” the tailor said, too quickly. “Really. It's just a needle.”
“But I feel terrible. I feel that I've let you down.” She stared up at him sadly from the haystack. She was so beautiful.
“You mustn't, you really mustn't feel so bad. Oh please don't. Even the most sensitive person on earh could be overwhelmed by the sheer numbers here. You see, as the number of pieces of hay (we'll call that h) increases compared to the number of needles, n, the probability, P, of finding n decreases monotonically without limit, and approaches arbitrarily close to zero.”

The princess felt something thumping in her own chest. “Wow,” she said. “You must be a poet.”
“No, a tailor. But please don't feel bad that you have failed.”
She sat up on the hay. “Failed? Not me. I just haven't succeeded yet.”
“But it's too hard a problem. Please, put it out of your mind. I wouldn't want to be the cause of...”

“Nonsense. We'll simply reconstrue. The problem, to find a needle in a haystack, is clearly solvable, even without a princessly backside. All we need to do is to move the h to one side, and everything that is left is n. While h is admittedly large, it is not infinite. The formulation of P, n, and h, which you put so prettily, is static, as it has no temporal element. If we consider instead the dynamic parameter P(t), the probability of finding n among h within a time t, then P(t) increases monotonically toward 1.0 as t increases.”

“Yes, but what a t. We're talking eons here,” the tailor said glumly. They might be ancient before they could ever get on to the love part.
“So now we reconstrue again.” She smiled contentedly at what was about to come, a bit of female logic to finish up the game. “Where you see here a haystack of order h, I instead see a needlestack of order n. Lost in the needlestack are h hays.”
“But n is still one and h is still immense!”
“True, but now suppose we search for the h instead of the n. We find the hay lost in the needlestack. Now the numbers are working for us and we're bound to succeed.”
The tailor breathed a sigh of relief. “Why couldn't I have seen it that way? What a loser I am.”

The princess ignored him and lay back against the hay. “Hello, I've got it,” she said at once. She reached under her waist and produced a perfectly splendid piece of hay. “One h. I have succeeded after all. We have succeeded. And now we can get on to other things.” She looked up at the tailor, who was rather cute, in addition to being poetic.

But the tailor was staring down at something glittering in the hay. “My needle,” he said. And he picked it up triumphantly.


He is a loser, she thought. Can't find a hay in a needlestack, even when the odds are stacked in his favor. How dare him to stop this continuous improvement plan so perfectly designed?

She shrugged and went on her way and he never saw her again.

This text is extracted from the book: 

Slack. Getting past burnout, busywork, and the myth of total efficiency
Ed: Broadway Books. Author: Tom DeMarco.

Click here to read the Spanish version of this article
Click the label English to see the other articles written in English


No hay comentarios:

Publicar un comentario