Ornate ballrooms have been putting giant mirrors along their walls ever since Versailles. Standing at the podium in one such ballroom, I admired how the mirrors created the illusion of depth, and effectively doubled the number of chandeliers hanging over my audience, as well as the size of my audience of corporate middle managers. I had just finished a forty minute talk on how to manage more effectively for shareholder value and was ready to take questions. A man with a bushy gray mustache stood up and took a microphone. He was shaking his head and looking rather disappointed.

“Frankly, I don’t get all this fuss about the shareholders. Really, who cares about them? Isn’t all this shareholder value b.s. just about enriching our top brass and those greedy bastards on Wall Street? What about the workers, communities, and other stakeholders?”

My experience told me that hiding corporate waste behind “stakeholder” interests was popular sport among many managers. “Screw the shareholders” was the unspoken code across a wide swath of corporate America. My first instinct was to sternly insist that any subordination of shareholder interests to those of other stakeholders is legally improper, morally indefensible, and financially suicidal. But I took a moment to consider a less polemical response. I stared off to the side to avoid the audience while I tried to compose a thought. But the audience was still there, in those giant wall mirrors. And there I saw an answer.

I looked back to the gentleman with the mustache and said, “You’re right about your bosses. They’re in it for themselves every bit as much as you are. And those hounds on Wall Street, you’re right about them, too. You know, I go out drinking with those guys–fund managers, investment analysts. All they care about is seeing the money. But I’ll tell you, they’re also driven by a deep-seated fear. That’s right, they’re scared. On their third or fourth margarita, they’ll tell you plainly what keeps them awake at night. It’s this group of folks so ruthless, so relentlessly avaricious, so impatient and demanding…” I said shaking my head theatrically. “Hey, wouldn’t you know it? There they are!” I pointed to the left wall.

Everyone turned and saw themselves in the large mirrors. Soon, they saw themselves laughing.

I continued. “Over sixty percent of equity investments in this country are owned by people like you, mostly through pension funds invested on your behalf. If the person managing your retirement money told you that you earned a five percent return last year when the overall market earned 25 percent, how would you react? Hey, no problem, right?”

Many in the audience, including the man who asked the question, nodded as if to say “yeah, right.”

Red Auerbach was a heck of a coach. His Celtics won eleven NBA titles in 13 years. The dominant Chicago Bulls of the ’90s would have had to win another five straight championships to match Auerbach’s record.

Red didn’t like metrics.

Like all great coaches, Auerbach had a mental framework for what it took to win at his game: The person with the best shot should take the shot; the team should ensure that that person gets the ball; the team should prevent their opponents from getting good shots. Pretty straightforward.

Still, most accounts of basketball greats focus on their individual stats–total points, points per game, rebounds, etc. Auerbach didn’t lean on that quantitative crutch. He believed that none of those stats spoke to the true value of his players.

“There’s only one stat I was ever concerned about,” says Auerbach. “When this guy’s in the game, does the score go up in our favor or go against us? The Boston Celtics never had a league’s top scorer. We won seven championships without ever placing one Celtic in the top ten.”

This wasn’t an accidental result. “No Celtic got rated according to how many points or rebounds or assists or anything else he might have compiled. Each man was assessed according to his contribution toward making us a better team. That’s all I cared about. In our system, the guy who sets the good pick was just as important as the guy who made the shot.”

How does one measure that? Auerbach measured it by watching it. In practices and games, “I saw what I saw,” and he recognized and penalized lack of teamwork and false hustle.

Now, imagine his coaching task if, before every game, he went behind a curtain and could only manage his team by receiving stats and the occasional play-by-play? Here, the inherent limitation of metrics becomes quite plain.

So, why does someone who sells metrics for a living offer such a limited view of them? Because a business unit is distinguished from a sports team by organizational complexity. Unlike coaches, business managers can’t watch the whole game as it’s happening. They can take little peeks from behind their curtain to see part of the crowd, a few players on the floor, maybe a ref for a chat. My clients benefit from maintaining the distinction between the measurement of “key factors” driving value, and the rich reality of decisions, actions, and environmental responses that actually create value. Anyone practicing metric madness in the belief that they can eventually get enough data to manage an organization should try coaching a competitive sport while remote from the action. Metrics are a tool for management, not a substitute for it.

Having said that, one measure does ultimately count. Every organization has a goal that happens to be represented a hard number. In Auerbach’s case, it was wins and losses. In business, it’s how much money the company actually makes over time (or, the value of the firm, in PV terms). Oddly enough, many managers drowning in a sea of metrics often fail to keep that one, ultimate goal in mind.

A few key metrics can give you an excellent idea of where to look for competitive advantages or shortcomings, but then you have to look. That’s why great managers always supplement customer, employee, and productivity measures with constant visits to their customers, people, and operations. They know the value of a first hand feel for what’s driving their business, and usually demand the same from their direct reports.

And they keep a close eye on the score.

One of my clients came up with an unusual angle on what he felt I was aiming at in my pursuit of a better measurement system for his company.

“We could have like this box, say, in a corner of the office, that indicates the value of everything you decide to do. Like, I’m meeting with Jack in engineering, that’s going to be worth $500. Or, I’m going to the bathroom, which would be worth minus $20. Unless, of course, I really needed to go, and my productivity was beginning to suffer.”

It was a wacky notion, this magic value box, and it had a certain intuitive appeal. But I suggested that such a box would really be more of a distraction than a panacea.

“Consider what you’re asking this box to do,” I said. “It has to tell you the likely contribution to the value of your whole firm from some individual activity. This would be mathematically impossible for things more complicated than turning out the lights when you leave.” I reminded him that major drivers of value, like strategic positioning or operating efficiency, are team efforts that don’t lend themselves to being objectively disaggregated to the individual level.

So this manager suggested, “then we need a box that gives such a value for team efforts rather than individual efforts.”

“Maybe, but that’s still like driving using the rear view mirror,” I replied. “The box you’re describing works, like any measurement system, only when you take some action. But there’s limited benefit to measuring, even in real time, the value impact of a particular decision after you’ve acted. The big benefit comes from understanding which choice from a range of possibilities will generate the greatest value before you act.”

“So,” he continued somewhat less excitedly, “we need a predictive box instead of a reactive box. One capable of modeling a nearly infinite variety of possibilities, and efficiently selecting the one likely to create the most value, and in real time.” I nodded. He looked stumped. “Well, I guess that sounds impossible.”

“Actually, you already have one of those,” I said.

“I do?”

“Yeah. Your brain.”

My clients learn from me, and I learn from them. They stretch my understanding of what we’re trying to accomplish, and my ability to articulate key ideas, like the distinction between the tracking versus incentive value of measurement.

I’m often reminding my clients that the right measures only get you part of the way there. It’s accountability for results that gets one’s brain into gear. An objective, value-based standard not only helps us identify which problems are worth solving, or enables us to efficiently cull the “nearly infinite variety of possibilities,” but also compels us to use our imagination, energy, and courage to solve them. These solutions often involve difficult and complex trade-offs — the kinds worthy of that “magic box” on our shoulders.

I was called in by the management of a giant telecom whose finance department was in a tizzy. Their business unit managers were apparently engaging in sale-leasebacks of major facilities in order to grow their economic profit. Their motivation was obvious—their bonuses were based on EP. As these operating managers improved their bonus outlook, however, each of these complex transactions destroyed a little value for the corporation. Finance was appalled.

“So,” I said, trying to understand, “the last time I was here, helping you design the roll-out of this measure, you were telling me that your ‘Bellheads’ would have trouble with this. You were skeptical that they would be able to use the measure to change their operating behavior. And now, these guys have figure out how to arbitrage your cost of capital. Is that it? Yes. And are you having any other problems your EP-based bonus plan? No. OK, then.”

When I checked with the operating managers, they described these EP-boosting deals with pride. What looked to the finance department like a loophole needing to be closed was, to the business unit management, a clever strategy for boosting the measure for which they were held accountable.

I reported back to Finance. “You have basically three choices. You can redefine the cost of capital to account for these deals. This would make your compensation measure difficult for your managers to game, but also more difficult for them to manage. Or, you could centralize lease vs. buy decisions, allowing you to enforce better discipline on them, but also adding a bureaucratic layer to business decisions. Or, you can let it go, which would be my choice.”

The finance managers gasped. What kind of consultant faced with a clear problem tells us to do nothing? I told them to think about it. After a week of thinking and discussion, they seemed better able to grasp how significant were the cost of complexity or the cost of intervention. True, the sale-leasebacks bled hundreds of thousands of dollars per year. But undermining the authority of business unit managers or slowing down their decision making could hurt the company far more than that. So would undermining the integrity of their compensation plan with changes specifically designed to reduce their bonuses. The company eventually accepted my suggestion, with Finance educating the business units on the proper evaluation of buy vs. lease decisions, and having those units report their leases in a timely fashion so that they knew their decisions were getting at least some scrutiny.

The number one reason I am given for why a company drops an EP or EVA plan is that it’s too complicated. Indeed, poor implementations of EP-based plans often yield unnecessarily complex metrics, which can lead to many problems. But I have repeatedly seen non-financial managers figuring out complex strategies to make a few extra bucks. I have also seen plans dropped because they didn’t pay out because the performance wasn’t there, and still heard “the measure was too complicated” getting the blame. One struggles to find the right balance of empathy and skepticism in dealing with management on the things that affect their pay.

Item 1 from the Wall Street Journal, January 12, 2004

“Shell Lowers Proven-Reserve Estimate by 20%

“Because of Shell’s disappointing results, Sir Philip has been under fire almost since taking over the top Shell job in 2001…The reserve issue has re-ignitied questions about Sir Philip’s future, especially since he led Shell’s exploration business for much of the period that the company said it overbooked reserves.”

Item 2 from the New York Times, January 31, 2004:

“Coke Employees Are Questioned In Fraud Inquiry

“According to two former employees who have been interviewed as part of the investigation, the federal investigators are particularly interested in the role of the company’s chairman and chief executive, Douglas N. Daft. He ran the company’s Middle East and Far East businesses in 1999, when the two employees say the channel stuffing took place.”

Is it worth noting that the main metric for Shell’s exploration business is ‘reserves’, and for Coca Cola’s international business it’s ‘volume’? The prediction here is that Watts and Daft are out in a month, but the connection of these dots may take a year or two.

As a young manager in an old transportation company, I learned the distinction between a good story and a meaningful explanation, and that one rarely leads to the other. It began when my CEO periodically asked me this question that, on the face of it, made no sense: “What drove the two percent (or whatever) profit growth in your business last quarter?” He was looking for specific business segment data. So I would identify some major pieces of my business whose net growth accounted for the two percent gain, leaving out the single or double digit gains or losses of my other nineteen segments that also contributed to that net two percent. It was a meaningless answer passing for analysis, and I knew it, and I suggested as much to my boss. He smiled and said not to fret. These numbers were for the consumption of reporters and analysts. “They’re just looking for a story,” I was told.

Of course, we all endure such “stories” daily from the business media. They pass these stories off as meaningful explanations for a company’s results, or a sector price movement, or even why the whole market rose a percentage point. But markets are granular. An endless permutation of select offsetting factors can “explain” your typical percentage change, which drain the nominal explanation of all meaning. We know they’re meaningless because if someone gave us the story facts beforehand, we’d have very little edge in predicting the outcome those facts were meant to explain. I can’t tell you how many times I’ve seen analysts warn of something, like the peril of rising interest rates for a particular stock, only to see interest rates go up — along with the stock price. I’m not suggesting that interest rates weren’t relevant to the value of the company, only that it was obviously far from the whole story.

Consider the extreme case of a company’s major plant hit by an aircraft and destroyed by fire. We shouldn’t be surprised to see its stock price drop. We would say there was a strong causal link. But what if the stock price goes up, say, four percent? Maybe the plant was well insured and otherwise in dire need of modernization.

Or consider a company announcing that it will restate two years worth of financial results because of a change in the way it accounted for several transactions, dinging earnings for those years. Furthermore it will also announce that earnings in a future quarter will be lower due to reduced capital spending. When it’s stock price drops, the “earnings” story would seem like a slam dunk. Unless the stock price jumps fourteen percent.* Then the story becomes more complicated — the market already discounted all the bad news, or saw the restatement of earnings as a proactive step toward making their accounting more transparent, or thought that the reduced earnings were more than offset by reduced capital spending. That’s the problem with stories — most of them are accurate; none of them are true.

Company value is what chaos theorists call an emergent phenomenon. It emerges from a complex reality that outsiders can rarely fathom. Talking about results as the product of this or that effect is cheap and easy. A manager accountable for those results must work from a much better model. She can’t afford to believe her own stories, or have her judgment replaced by those looking through the lens of their stories.

Fortunately, most managers have the training and experience to deal with that little portion of the real world whose complexity is transparent only to them. The best you can do is hold them accountable for results. By the time you read about the results in a release, or an analyst report, or in the paper, when you’re getting maybe two percent of what really happened, it’s hopefully a positive two percent.

* “Hanover Compressor to Restate Results,” Wall Street Journal, February 27, 2002