Monday, October 26, 2009

A poka yoke for strategy

Tom Jackson
Portland, Oregon
26 October 2009

What is hoshin kanri?

It is, essentially, a way to create a new organizational structure.

What is organizational structure?

According to economists like the new Nobel Laureate, Oliver Williamson (see my previous post), organizational structure is a structure for processing information. This makes complete sense if we stop but a moment to consider that organizational structure is about who makes the decisions. And decisions, as we know, require information and a set of priorities. Voila! Information processing.

So what kind of organizational structure does hoshin kanri create?

Simple. It's a poka yoke for strategy.

Come again?

A poka yoke. A mistake proofing device.

To quote Shigeo Shingo, a poka yoke or mistake proofing device "builds the function of a checklist into the process." Shingo illustrated his concept as a system of nested Deming cycles. (Actually, Shingo used the more economical Plan Do Check cycle, but never mind.)

Here we see how it works. We start a Deming cycle or cycle of organizational learning. To ensure flawless execution, in the very act of DO-ing, we embed another learning cycle. We do this for two reasons:

1. To be certain that we are doing it the right way, i.e., according to standard work; and
2. Just in case the "right way" isn't good enough, because you never know....

We can add checklists to checklists ad nauseum, depending only upon how much risk we want to wring out of any given system. The structure is sometimes called a nested Deming cycle. It's a checklist within a checklist within a checklit....

In military craft we refer to this design strategy as redundancy.

Back to checklists. Think of checklists as a form of memory. This is very much like computer memory.

In the case of operations, the checklist is about not forgetting how to do a job right the first time. The checklist can be literal (as in an airline pilot's checklist) or physical (as in a rumble strip or the alarm that is triggered when you leave your keys in your car).

In the case of strategy, the checklist can be literal, but it is essentially a metaphor for a system of constant reminding and recalibrating. Through the process of catchball, we agree to return to the "checklists" or action plans that we built into our A3 team charters, when we first built our annual hoshin, our strategy, our policy for improvement. Moreover, we remind each other. And when things don't work out quite as we had originally strategized, we regroup, refocus, and on we go.

Once a month, once a quarter, once a week (and by extension, once or more every day in our standup meetings), we revisit our "checklist."

Are we in control or out of control.

It is easy enough to see how this might apply to daily work, but to strategy? Well, why not? And that is precisely what hoshin kanri does.

The example of pilot and co-pilot double-checkling the aircraft before pushback comes very close to capturing the spirit of hoshin. The pilot checks the plane and checks the co-pilot's checking of the plane. Likewise the co-pilot checks the plane and checks the pilot's checking. Top down; bottom up. This is the essential structure of every hoshin review meeting.

And the checklist?

The organizational "checklist" is the list of priorities, as expressed in the organization's mission and vision and in every A3 team charter drafted during the catchball process. (As every game theorist knows, strategy is one very big checklist.)

Hoshin kanri is about remembering what we all agreed was important and what to do together to promote the interests of the organization. And if we should forget, or if circumstances should invalidate our agreements, hoshin is about starting over (and over) again, remembering what went wrong, what went well. Hoshin is about honoring the very human endeavor of nothing ventured, nothing gained.

Trial by fire, the fire of intelligence.

Thursday, October 15, 2009

Oliver Williamson

Although I have never met Professor Williamson, who won a Nobel Prize for Economics this week, I have long been under his influence. I was very broadly educated in the New Institutional Economics at Indiana University (no, I didn't study with this year's other winner, Professor Elinor Ostrom, of, oh yes, Indiana University), where I was tutored by the late Professor David Martin in Williamson's fine theory. Incidentally, in the year before Williamson switched from Yale to UC Berkeley, he spent at year at I.U., which attempted to recruit him, unsuccessfully. (Alas, I had already graduated the year before.)

Williamson's theory goes something like this:
  • Business firms exist to minimize the costs of doing business, called transaction costs, which are the costs of searching for good partners, negotiating contracts, and then enforcing them.
  • Although we would like to find business partners who are not difficult to find, easy to get along with when we do, and who routinely live up to their agreements without being bothered by litigation, human nature is such that this is not always, or even often, the case. Human beings have very specific, often severe limitations, among them: a) small slow brains that can only handle several (7) chunks of information at a time; and b) sneakiness, that is, a tendency to withhold valuable information when it will benefit us individually, and to heck with you.
  • Given this unfortunate situation (our lack of mental acuity, and/or that of our potential partners--and the fact that everyone is out to get us) the enforceability of contracts is a really big deal. Especially when assets are highly "specific," as they often are. "Specific" means specific to the contract. An example of a "specific" asset is, say, your construction of a large mine in a foreign country--a very expensive proposition. And, let's say, the government of this country decides to nationalize the mine. You lose your shirt. Because, your shirt was specific to the contract. In other words, the value of the asset depended upon the participation of the other party, in this case, the government.
  • Or, closer to home, consider the case of other people "taking credit for your work." Let's say that you are a functional manager in possession of specific information about a process. Or let's say that you have a brilliant idea that you need help to realize. The value of that information or idea may well depend upon the willingness of manager's from other business functions (in the first case) or a joint venture partner (in the second case) to cooperate in pursuing a common goal. Your information or idea is a "specific asset" in the contracts (largely informal, in many many cases) that you make with your fellow managers or joint venture partners.
Okay, what does this have to do with hoshin kanri and the A3 process?


Hoshin kanri exists to minimize the transaction cost of doing business, and Williamson's theory is a near perfect guide to explaining how, and why, hoshin works.
  1. The PDCA process at the heart of hoshin speeds up our small slow brains, courtesy of the scientific method--PDCA effectively offsets our unfortunate lack of gray matter;
  2. The catchball process of hoshin is slow and deliberate, which levels the playing field between mental tortoises and mental hares (this has other benefits, because sometimes the slow are deep)--on the one hand catchball is a very effective search process, and it makes it more difficult for the sneaky to take advantage of the honest;
  3. The use of A3 documentation (an A3 is essentially a contract--A3 team charter--"charter" is another word for "contract") in the catchball process makes it very difficult, if not impossible, for sneaky people to withhold vital information--the process of hoshin kanri is transparent and this rather effectively offsets the human being's natural tendency to be sneaky;
  4. Hoshin's reporting system of short, frequent, focused meetings again makes it difficult to withhold vital information. Also, individuals who are not compliant with the terms of their "contracts" are discovered almost immediately, and encouraged to comply;
  5. Finally, compliance is further encouraged by hoshin's drive-fear-out-of-the-organization approach, because noncompliant participants have relatively little to lose as well as little to gain from gaming the system.
There are other, more subtle ways, in which hoshin kanri and, indeed, the traditional tools of lean operations reduce the transaction costs of doing business. Some of the more subtle methods address the costs of searching for information. Take, for example,
  • Standard work
  • Successive and self-checks
  • 5S and visual control
  • Poka yoke or mistake proofing
All of these methods are designed to bring "defects" (which, because of standard work, may be understood as "breaches of contract") to the surface. Keep in mind, according to Williamson, transaction costs are the costs of:
  1. Searching for information;
  2. Negotiating contracts, formal or informal; and
  3. Enforcing contracts.
So efficient is hoshin kanri's system of achieving organizational alignment and encouraging members of the organization (including those pesky suppliers!), that I and others have spoken of Toyota's marvelous system as a system of self-organization. Another word for this is cybernetic. Toyota is a cybernetc or self-organization system.

You've had enough? Williamson's theory is about as clear as mud. And reading his wonderful books only makes it worse! But his theory is correct. It rests upon a profound understanding of human pyschology and markets that has been awarded the Nobel at least twice before (to Herbert Simon and Ronald Coase, with whom Williamson studied). And, if you really, really want to understand what makes Toyota tick, you need Williamson's theory.  Maybe one day, after I put my darling daughter through college, I'll write a book about it. It explains Toyota's miraculous supply chain, too.

In the meantime,

Congratulations, Professor Williamson!

Thursday, October 8, 2009

Ferrari + hoshin kanri = lean enterprise

As I watch Toyota's leadership go through its latest bout of hansei (deep reflection), I am struck again by how conservative Toyota is. Compare Toyota's apology for it floormats to how Ford swept the Pinto's exploding gas tanks under the carpet!

Is this profoundly conservative attitude rooted in Japanese culture? Well, maybe. Or maybe it's rooted in modern portfolio theory.

All of us have retirement accounts and most of them are managed by brokers who have been trained to do one of two things, depending upon our appetite for risk. 1. Our brokers can maximize the return on our portfolio for a given level of risk preference; or 2. They can minimize the risk that we will fail to achieve a given rate of return.

When we are young masters of the universe, we tend to prefer risky investments that maximize our return, because if we experience losses we still have time to start over. But our appetite for risk diminishes as we get older and acquire more obligations to family and business partners. At some point we may pile up enough money and then we ask our brokers to minimize the risk for a given rate of return.

So what does this have to do with running a business? We can easily view a business organization like Toyota as a portfolio of "real" (as opposed to financial) "investments" in people, processes, and relationships. Just like stock broker, what a business does with those investments is governed by the risk/return tradeoff discovered by Markowitz in the 1950s.

At the same time, however, this is inconsistent with the microeconomic theory of the firm, which states that businesses maximize profits. What gives?

I'm going with Markowitz.

Toyota, it seems to me, behaves like a good family man--maybe even like a grandfather. Toyota is not a profit maximizer! Instead, Toyota is a risk minimizer. Indeed, properly understood, the 7 Deadly Wastes are sources of risk, and that is why Toyota strives to eliminate them. And the method of hoshin kanri? It is essentially a method of portfolio selection and management.

That explains Toyota's dull styling. Or does it? Actually, even Ferrari can be explained by the same theoretical idea. Ferrari is the young man who stands in contrast to Toyota's grandpa. Both are governed by the risk/return tradeoff.

No one would ever call Ferrari lean*, but the analogy raise an interesting possibility. Ferrari could well employ lean techniques--in particular the technique of hoshin kanri--to tweak its portfolio of real investments. While Toyota uses hoshin to minimize the risk of hitting a target profit--and thus builds dull but reliable cars, Ferrari could take on more risk and maximize its return by building sexier cars. It makes sense, surely. If Ferrari did not take risks, we would not love their cars, nor would we be willing to pay so much for them! The question is: Precisely how much more risk can Ferarri take on?

Modern portfolio theory tackles just this sort of question. And the portfolio theory of lean enterprise suggests that there is an answer to Ferrari's question. The benefit to Ferrari of employing hoshin kanri would be to calibrate its risk/return tradeoff scientifically to please its customers as well as its bankers.

And, dear reader, if Ferrari did this, we might just have to call Ferrari a lean enterprise, despite having much more "waste" in its processes, relative to Toyota. Why? Because Ferrari would have optimized its portfolio of investments in people and processes to provide the best value for its very particular clientele. With higher risk, there will tend to be more waste. But there will also be more daring curves, more moving parts, more sex appeal. After all, Ferrari's customers are not shopping for a ride to the grocery store.

* Deming Prize-winner Ryuji Fukuda once visited Ferrari and offered to help them reduce rework, lead time, and cost. Ferrari's response was: Why? Fukuda opined that, "If there is a more expensive way to do something, Ferarri will find it."