Wednesday, September 30, 2009

How is a lean enterprise like a Picasso?

A cubist painting, such as Picasso's Three Musucians, pictured here, was intended to convey in two dimensions all of the essential information about a three dimensional object or event. Some observors saw something more, namely, a version of reality from multiple perspectives.

Picasso's Three Musicians

Consider Picasso's painting, Violin and Guitar,  and you can see for yourself how these paintings might make this impression.

Picasso's Violin and Guitar

Picasso and his cubist collaborator Braque denied the multiple perspectives theory of cubism. Never mind. I like the multiple perspectives theory, and apparently so does American artist David Hockney. Compare the cubist paintings here to Hockney's photographic collage of the Merced River in Yosemite Valley.

Hockey's  Merced River, Yosemite Valley

Now, my point about lean enterprise. Although we can plainly see in Hockney's collage that there are many different pictures taken--and assembled--at different angles, the overall impression is one of unity. Of course here we're dealing with artistic unity. In a lean enterprise, through the catchball process of hoshin kanri, we collect many more snapshots of reality that Hockney has ventured to put together. And the more widely we deploy our policy, the more perspectives we must integrate. Clearly, there is the temptation to forego catchball and simply dictate top down a nice, neat version of reality all tied together from a Renaissance perspective, focused on a "vanishing point" known as the CEO. Here we see Piero della Francesca's famous painting, The Ideal City.

della Francesca's Ideal City

And, clearly, this is how many companies are still run today, even those that claim to be lean. It is certainly how almost all ERP systems are designed! But note what's missing: INFORMATION!

In the Three Musicians, Violin and Guitar, and in particular in Merced River, there is much more information about what we are being asked to contemplate. In the Three Musicians we can see how the emsemble is literally intertwined. In Violin and Guitar, we can see the instruments from all angles simultaneously. And in Merced River, we literally see multiple perspectives on a landscape that would be impossible to integrate if the observer were to stand, as della Francesca, in a single spot. What's behind that round building at the center of the square. What's inside? All of this we could know if we let Hockney loose with his camera or Picasso with his oils and brushes.

Within a lean enterprise, because of the catcbhall process of systematically collecting the perspectives of senior leaders and all the managers--and in implementation the perspectives of all employees, we have a vision of the enterprise that is saturated with information, specifically, the lived experience and knowhow about the organization's processes.

When della Francesca painted Ideal City, mankind was undergoing a radical shift in perspective that gave birth to the individual and to the science of Gallileo, Kepler, and Newton. It is tempting to think that, with a shift in perspective so profound as we see in Picasso, Hockney, and, yes, the Toyota Production System, maybe mankind is on the threshhold of another breakthrough. Besides, we could use a breakthough right about now.

Tuesday, September 29, 2009

The meaning of profit management

Why is hoshin so powerful? People will tell you that it's the best way to get a company focused. No kidding!

But what does "focus" mean, apart from powerpoint slides that show all of our "arrows" pointed in the same direction.

It may be helpful to think of focus as the opposite of diversification. Diversification is what we do to minimize the risk in our portfolios of stocks and bonds. Diversification spreads the risk of losing on our investments as a whole by minimizing (at least based upon historical trends) the extent to which investment prices will go up or down together. We diversify because we are engaged in a game against the Market. To quote Peter L. Bernstein, the market is "a powerful opponent indeed and secretive about its intentions. Playing to win against such an opponent is like to be a sure recipe for losing. By making the best of a bad bargain—by diversifying instead of striving to make a killing—the investor at least maximizes the probability of survival." See Berstein, Against the Gods (New York: Wiley, 1996) p. 253.

Running a lean enterprise is in some ways the same as and in some ways different from financial investing. We can easily accept the idea the an enterprise represents investment in both real and intangible assets. The intangible assets of people and process are among the most important features of lean enterprises. And hoshin is the method by which those investments are chosen. But rather than diversifying our investments in people and processes, we focus. We can understand more clearly the meaning of focus by comparing it to the mathematics of diversification. Diversification requires us to minimize the correlation or covariance of the expected returns of investments within the portfolio. Focus requires us to do the opposite. To focus, we maximize the correlation of expected returns of investments in people and processes. Indeed, this is the whole purpose of "balanced scorecards" and the famous X-matrix.

Why would we diversify our portfolios of stocks and bonds and focus our portfolios of real investments in people and processes? The answer is: Our state of knowledge.

In the case of financial markets, we playing essentially against nature, the Market, about which we know only a little. The point to diversification is to avoid being taken by surprise. It's those surprises that prevent us from "betting the farm," i.e., from focusing on one stock a group of related stocks. In the case of lean enterprise, however, we know much more. Through QFD we know exactly what our customers want. Through hoshin we know our managers, employees, and suppliers rather well. We can afford to "bet the farm." In fact, to do anything less would be wasteful!

This becomes clear if we present this in mathematical form of modern portfolio theory, with a twist.

maximize COR (People, Process)

Subject to: Profit = X 

This says that we plan to maximize rather than minimize (that's the twist) the correlation among our investments in people and process, subject to a target profit, which we have arbitrarily set at X. In other words, we are going to choose investments in people and process that work together, the tighter the better.

This is new perspective on what Toyota calls "profit management." Normally, we think of profit management as consisting of its more familiar components, target costing of new products and kaizen costing of existing products. In most books about target costing and kaizen costing there is little discussion of hoshin and alignment (i.e., focus). That's a mistake. In this simple rewrite of Markowitz's portfolio theory, we can see what Toyota is really doing. Profit management really means that we are minimizing the risk of failing to hit a target profit. 

Moreover, the way in which we minimize that risk is to focus our portfolio, i.e., "bet the farm," on the most promising combination of people and processes, the combination that will deliver value to our customers. You've heard it before: Focus on the process, not the result. The result is but a constraint. It is given. It is the result of focusing on the right combination of people and process.

One last insight: Toyota is not a profit maximizer, at least in the short term. The company is clearly too risk averse. Like Markowitz's portfolio theory, this theory allows companies to maximize their returns (or profit), subject to some predefined tolerance for risk. In that case, the profit term becomes the "objective function," and the correlation term becomes the constraint. But we know that Toyota doesn't behave that way. 

Saturday, September 26, 2009

Don Quixote, hoshin kanri, and tilting at windmills

Tom Jackson
Monterrey, Mexico

I spent the past week in Mexico, where I made a presentation about hoshin kanri and the A3 system to the Annual Shingo Prize Conference of Mexico. In preparing for my talk, I searched for a way to connect with my Spanish-speaking audience. I recalled my recent reading of Cervantes' great novel, Don Quixote. Of course it is only coincidental that there is an "x" in both "Quixote" and the "A3X matrix," (the X-matrix is the subject of many of my posts) and that "X" looks something like the windmill that Quixote tilts with to such devastating effect early in the novel. Or is it?

NOTE: You don't need to read Don Quixote to understand this post. I'll do the heavy lifting. But, if you're interested, there's a new translation of Don Quixote in contemporary English--by Edith Grossman--that you might want to take a look at.

So, here goes. Don Quixote is, among other things, a book about how we deal with multiple realities. The book was, of course, written after Gutenberg published the Bible in German, an act that, together with Luther's Reformation, split the Western world in two. In this newly fractured landscape, a landscape that has become only more fractured since Cervantes' time, the idealistic Quixote tilts at the windmills because, having read too many Medieval romances, he firmly believes that they are horrible monsters. Quixote's more realistic squire, Sancho Panza, sees things quite differently and tries in vain to warn his master of the impending doom originating not from the windmills but from the Don's fertile imagination. Two perspectives; two different versions of reality. Which is correct?

Throughout the book we proceed through multiple adventures, disasters all, with Quixote and Sancho engaged in the same conflict of perspectives. This much of the story most people know from the Broadway show and movie, Man of La Mancha. But there is more. This is of course the world of the printed word, of two Bibles (and more to follow), the world of what was to become the free press, a world in which one may share one's own perspective in one’s own language—although at considerable risk, in our own time as well as Cervantes’. Characters in the book start writing about Quixote and his squire; their fame, or should we say infamy, grows. Before the end, the whole world is reading about Quixote and his slapstick, tragicomic adventures. In fact, the new authors join in fun, actively creating new misadventures for Don Quixote and Sancho Panza. The dividing line between reality and illusion is blurred. The book culminates in Quixote visiting a bookstore in Barcelona, where he reads about himself, as seen by others, and he complains bitterly, "This is not the true Quixote." It all ends badly. Quixote dies unhappy, having seen his dreams of romantic glory come “true,” orchestrated by the new authors only to make cruel fun of the old man. But it all unravels. The multiple perspectives of the book cannot be reconciled. Everything falls apart.

Clearly, I do not do Cervantes justice, but that is not the point. My point is that this should ring a bell. Aren't many leaders a little like Quixote? Isn't leading an enterprise something like this? The advent of lean enterprise only makes the situation worse. There were always many perspectives to reconcile in leading a company. Now, in the era of empowerment, leaders have to listen to Sancho Panza as well as all of those imitation Quixotes! It is, indeed, a quixotic task. And perhaps we are as crazy as Quixote to believe in our visions, saddle our broken down horses, and stumble off to tilt at windmills.

Or else we resort to the unilateral assertion of authority. But we know that Inquisitions are ultimately not very effective. Besides, they are a waste of good human resources.

In the method of hoshin kanri, there may be a happy ending for the quixotic leaders of enterprises that would be lean. For hoshin kanri is truly a method for reconciling the multiple perspectives of the enterprise in alignment with the vision of the Don, er, the leader.

The catchball process of hoshin kanri is a painstaking methodology for sharing perspectives, perspectives on the targets and the means by which the targets will be achieved. Senior management communicates its vision and establishes the results that must be achieved. Middle management responds with comments on the vision and targets and proposes the means or improvement processes by which the results with be achieved. In turn, middle management communicates to front line managers the process improvements that will be undertaken, and front line management responds with proposed schedules and identifies resource requirements. 

When catchball is a genuine process, there is what contract lawyers call a meeting of the minds, which is the core of any robust agreement, the source of organizational alignment. When done systematically and methodically, catchball results, for all intents and purposes, in “one reality” for the members of the organization. We can see this unity of perspectives memorialized on a good A3X, or X-matrix, which binds together all of the A3T team charters that define the annual hoshin or short term strategy.

It is as if Don Quixote has convinced the entire world to join him on his quest.

Naturally, the Don—and by extension, all who would be leaders—must take care to envision New Worlds that are truly valuable, namely, worlds without waste, worlds that respect people.... Even here, however, hoshin offers a checking mechanism. The Don must listen to Sancho--and all of the imitation Quixotes--and each of take them seriously. 

Luckily--in fact, deliberately--Sancho, and the catchball process, ultimately tie the Don's vision to the gemba of direct obervation.

Monday, September 21, 2009

Hoshin kanri lowers the cost of doing business

Hoshin kanri lowers the cost of doing business.


The reason is twofold:
  1. Organizations are a nexus of contracts; and
  2. Hoshin kanri results in what economists call "complete" contracts.
Before we go any further, let me be very clear: A3s are contracts. Moreover, hoshin kanri is essentially a method for creating and enforcing A3 contracts. (The principal A3 is Toyota's Proposal A3, which is essentially a team charter, and "charter" is just another word for "contract.")

Now, read on.

Organizations are a nexus of contracts

...and, by extension, lean enterprises are a nexus of A3s.
  • nexus |ˈneksəs|
    noun ( pl. same or -uses )
    a connection or series of connections linking two or more things : the nexus between industry and political power.
    • a connected group or series : a nexus of ideas.
    • the central and most important point or place : the nexus of all this activity was the disco.

Economist and lawyer John R. Commons, teacher of the Nobel Prize-winning economist and psychologist Herbert Simon, pointed out that the fundamental unit of organization is the--you guessed it--contract. Business firms arise when the "external" or "spot" contracts of the free market are displaced by "internal" contracts between business owners (today's shareholders) and their managers, employees, suppliers, and of course customers. For a variety of reasons, including the standardization of tasks and the possession of an institutional history or "memory," business firms exist because they have a relative cost advantage over and above the free market. Namely, they reduce the "transactions costs" or costs of doing business. These costs consist precisely of the costs of negotiating and enforcing the contracts between shareholders, managers, employees, etc., that allow us to coordinate our behavior in a purposeful way.

Now this is very curious, for it suggests that business firms that can find a better way to negotiate and enforce the "internal" contracts associated with running a business will have lower costs of doing business than their competitors. And this has in fact been born out in at least two major cases. The first case was the Rise of Big Business in the mid to late 1800s, beginning with the building of the transacontinental railways in the United States, and again in the1920s with the invention of the modern corporation at General Motors and DuPont.

Followers of this blog will be aware that this has happened a third time, at Toyota.

Hoshin kanri results in "complete" contracts

...and, by extension, A3s are a better kind of contract.

Before we can explain what a "complete" contract is, we have to understand what is a contract. In its simplest form, a contract is a mutual promise between two or more parties, based upon a "meeting of the minds" between those parties.

So, what does it mean, a meeting of the minds? Let's say that you and I make a contract. When there is a meeting of the minds, it means that I understand and value what you have offered, and you understand and value what I have offered, and furthermore that we are both aware of this situation. In other words, there is a confirmed understanding of the mutual value we bring to the exchange. This will be true in every case, including the simple exchange of goods, as in the exchange of money for a pair of shoes, or in the more complicated case of "performance contracts" that involve the exchange of money (or some other valuable thing) for the performance of certain services, as in "I'll scratch your back if you scratch mine."

Enough about that. What makes a contract "complete"? Basically, a complete contract is a contract in which the meeting of the minds is without any gaps in mutual understanding.

Completeness is a concept that pertains to our state of knowledge about what game theorists call "the rules of the game," in other words, our knowledge of each other, our intentions, capabilities, understanding of the world, and modes of communication. There are several ways in which a contract can be incomplete. For example,
  1. We can be uncertain about the true identity of the parties to the contract;
  2. We can be uncertain as to the true intentions of the parties;
  3. We can be uncertain about the capabilities of the parties to fulfill their obligations as specified in the contract;
  4. We can have different versions of history or "reality" of which we not are fully unaware;
  5. We can have different languages (or codes) and channels of communication of which we are not fully aware.
That's a lot of information to track. And tracking information is a pain in the the patoosh. So, clearly, there are costs associated with incomplete contracts. These costs are the costs of guessing about who the parties are, what they really want, what they can actually do, what is their version of reality, and how they mean when they attempt to communicate with us. Another way to talk about "guessing" is to talk about "searching for and processing information" to "make decisions" (the topic for which Herb Simon won his Nobel Prize).

Game theorists model the "incompleteness" all of this guessing by imagining that decision-makers generate multiple, interrelated probability distributions concerning each category of uncertainty.
  • Fans of the movie, A Beautiful Mind, may be aware that economist, John Nash, played by the crazy and brilliant Australian actor, Russel Crowe, the movie's brilliant, crazy (did I mention brilliant?) protagonist collected his Nobel Prize with two other brilliant game theorists, John Harsanyi and Reihardt Selten, whose theory of completeness I am expounding here.
Multiple probability distributions? That's a lot of mental bookkeeping. Crikey! Obviously, mental bookkeeping does not come without a cost, the cost of gathering and processing information, the cost of "guessing." The cost of doing business...

So, how does hoshin kanri create more complete contracts?

In a word, "catchball."

As I have written in my book, Hoshin kanri for the lean enterprise, catchball is a process of negotiating A3s or internal contracts. And what is negotiation but a process of testing and confirming the identity, intention, capability, understanding (and, as necessary, communication modality or style) of the parties to the contract?

Moreover, the catchball process is a process of the careful communication and confirmation of contract particulars. In the literature, the particulars of internal contracting are often referred to as the "whats" and "hows" or, in other words, "targets" and "means," respectively, of contracts between managers and their direct reports. In a mature lean enterprise, there is very little "contract ambiguity," i.e., little misundertanding among managers about what is to be done and how it will be accomplished. Likewise, there is little misunderstanding about how the parties communicate with one another, because that is highly specified through the hoshin process of "check." (This is a topic for a future post.)

Cut to the chase

Because of hoshin kanri's catchball process, A3s--the definitive internal contracts of the business firm--are highly specified, i.e., much more complete, and therefore there is a lot less guessing in the lean organization. Less guessing about what is to be done and how it will be accomplished. Less guessing about who's on first and what they may be thinking.

Less guessing implies lower costs of doing business.

Hoshin kanri implies less guessing and therefore implies lower costs of doing business.

Lower costs of doing business imply greater competitiveness.

Therefore, companies that implement hoshin kanri will be more competitive.

What are you waiting for?

Tuesday, September 15, 2009

A Balanced but Disconnected Scorecard

Twenty years ago, Robert Kaplan of the Harvard Business School and his business partner David Norton witnessed Analog Device's implementation of hoshin kanri and the result was the Balanced Scorecard. Of course the Balanced Scorecard was wildly popular and made many consultants a lot of money. But, the Balanced Scorecard is not hoshin kanri. This remains true despite Kaplan and Norton's dogged efforts to turn the Balanced Scorecard into a genuine strategic management system. To their credit, they have done a nice job integrating the business school literature on resource-based competition, and for this important task we thank them. But, every time--in the course of my lengthy lean manufacturing and lean healthcare consulting career--I run into it, I am always struck by the fact that the Scorecard never makes it far from the board room. I have never seen it connected to daily work on the front line. What happened?

My hypothesis about the Balanced Scorecard is that, despite Kaplan and Norton's good intentions, they didn't understand what they were looking at when they examined hoshin kanri. Perhaps Analog Devices didn't get it right in the first place, something that a short trip to Toyota would have cleared up. (This is the route that H. Thomas Johnson and Robert, old colleagues of Kaplan's, both took.) 
I believe that Kaplan and Norton made two errors:

  1. The first error was to fail to understand that hoshin kanri exists mainly to make systematic adjustments to standard work. To be effective, a balanced scorecad must somehow be connected to the mechanism by which the drivetrain achieves traction. Strategic traction can only be achieved if the company's leaders make contact with the people on the front line of the organizaiton, where real value is added.
  2. The second error was to fail to realize how thoroughly--no, how radically--decentralized Toyota truly is. With adherence to standard work assured through 5S, visual control, and poka yoke, things are pretty much under control at all times. And, given that standard work incorporates all elements of cost (task, sequence, time, and work-in-process inventory [staffing can be derived from task, sequence, and time]), this means that adherence to standard work amounts to auditing in real time. This is why Prof. Johnson reported that Toyota does not permit accountants on the shop floor.
The upshot of these two errors is a balanced scorecard that can be retrofitted to the pre-Toyota corporation, which was (get out your history books) invented at General Motors in the 1920s. The original "balanced scorecard" was not designed for a retrofit; it was designed for a double kaikaku of corproate structure:
  • Kaikaku 1: Decentralize radically by establishing standard work and empowering front line employees to stop the process to fix problems;
  • Kaikaku 2: Enable the matrix organization by abandoning functional silos in favor of cross-functional and interorganizational teams.
This is in fact the whole purpose of hoshin kanri and any balanced scorecard worthy of the name. 

Tuesday, September 1, 2009

Optimization, nonprice factors, risk, and the balanced scorecard

In my last post I argued in favor of optimizing returns in business by minimizing the risk of failing to meet quality, cost and delivery targets. And to blazes with profit maximization.

This post is an addendum. I wish to point out that my position is already implicit in the so-called balanced scorecard. (My readers will know that the balanced scorecard is largely derived from hoshin kanri.)

The balance in the balanced scorecard arises from the inclusion in the normal profit maximizing calculus of nonprice factors such as quality and delivery (from the so-called process perspective) as well as people development (from the so-called growth perspective).

One can easily extend the notion of "balance" to include risk, as demonstrated by McKinsey and Conference Board research on ERM (enterprise risk management) (links to follow in a subsequent post). See, e.g., Kevin S. Buehler and Gunnar Pritsch, "Running with Risk," McKinsey Quarterly, Winter 2004, pp. 7-12, and Carlyn Kay Brancato, "Enterprise Risk Management System: Beyond the Balanced Scorecard," Conference Board Report No. E-0009-05-RR.

The inclusion of nonprice factors already pushes us beyond the confines of profit maximization, at least in competitive markets. The inclusion of risk, however, moves us quite firmly into the world of optimization. We are forced to open price theory's "black box" and ask, "How to leaders really think," instead of adopting the mathematically convenient and, one might argue, socially disastrous convention of profit maximization.

-- Post From My iPhone