Dashboard paradise

Citation

Marren, P. (2011), "Dashboard paradise", Journal of Business Strategy, Vol. 32 No. 4. https://doi.org/10.1108/jbs.2011.28832daa.003

Dashboard paradise

Article Type: Alternative strategies From: Journal of Business Strategy, Volume 32, Issue 4

Patrick MarrenConsultant at The Futures Strategy Group, Crystal Lake, Illinois, USA

Strategy is about making decisions. Decisions are choices between alternatives. So strategy is, or should be, something that helps you make decisions.

“Dashboards” were all the rage a few years ago in the C-Suite. Probably still are, in some places. Dashboards are real-time displays of your company’s critical numbers, displayed on the desk of the CEO or COO. I have never actually seen one that was being used. I imagine that it would sort of be like tuning your cable TV permanently to the local weather update channel, except without the smooth jazz soundtrack.

The idea of a dashboard is very seductive, because it seems to promise to reduce a CEO’s or COO’s problems to a very few key variables. In practice, I bet, it is disappointing. Say you are sitting there listening to the smooth jazz and suddenly there is a beeping like you hear when a tornado warning is suddenly in effect for your weather area. One of your key variables is out of whack. Why? The dashboard probably cannot tell you. You may be able to query it, but all that might result is disgruntlement, ill-tempered phone calls, and something duodenal in your future (and your direct reports).

The problem is that while the dashboard tells you that you need to be concerned, it may hide the ultimate reasons why you need to be concerned. Why does it hide them? Because, in a sense, that is its function. You want to avoid wading through the data deluge and the complexity, right? You want to achieve a Zen sort of calm in the C-Suite, right? Well, certain things have to be cut out then.

Focus is great, and a dashboard is in principle a useful idea. You DO want to focus on the small number of variables that truly matter. The problem is that in a chaotic and dynamic competitive environment, the variables that matter are going to change all the time. Strategy will not show up on the dashboard – at least, it cannot be captured for very long on that dashboard. Strategy is not about monitoring key variables. It is more about choosing which variables matter.

I got an idea for a different sort of dashboard from one of the most important and least-well-known scientific achievements of the twentieth century (or any other): the simultaneous discovery of the mid-oceanic spread of tectonic plates outward from central ridges, and the previously unknown, geologically frequent reversals of the earth’s magnetic polarity. These two phenomena were both discovered some 50-odd years ago when a US Coast Guard vessel was sent for entirely unrelated reasons to tow a magnetometer across the ocean to map the sea floor.

Unexpectedly, the magnetometer recorded sudden flips in the magnetic properties of sections of the sea floor. Broad vertical (north-south) swaths of the sea floor would register a magnetic orientation in a certain direction; then, as the Coast Guard cutter continued across the ocean in a horizontal (east-west) direction, an abrupt boundary would be crossed, and the polarity of the next north-south swath would be precisely the opposite of the previous swath.

Scientists realized that the ocean was spreading from the Mid-Atlantic trench in both east and west directions, due to volcanic action in the trench, pushing the continents apart; at the same time, the lava pushed out from the trench in both east and west directions, in its initial liquid state, contained iron, which, due to the liquidity of the lava, was free to orient itself magnetically according to whatever the earth’s magnetic field happened to be at that time.

From the obvious pattern of spread from a central mid-Atlantic volcanic trench, scientists realized that the theory of tectonic plates, previously dismissed by most geologists, was quite real. And from the sudden abrupt reversals of polarity in the long-since-frozen swaths of undersea iron-rich rock, the scientists realized that the earth’s magnetic field had reversed itself frequently over its geological history. Two great fundamental discoveries about the basic aspects of life on earth, all thanks to a Coast Guard cutter sent on an unrelated mission.

So what does this have to do with anything business-strategy-related? Well, the sudden flips in direction experienced by the magnetometer put me in mind of a different kind of machine. Let’s imagine a slightly different sort of “dashboard” – a purely fictional one.

This dashboard would help you make decisions. You would feed in information about the competitive situation, and a dial would swing around between alternative courses of action. Let’s say, for the sake of simplicity, there are two possible alternative actions – proceed with the launch of a product, or not.

So you feed in data about your product and your competitor’s product. The dial whirrs around and settles on one alternative action or the other. Based purely on the relative attributes of your products, your “dashboard” tells you that you should launch your product.

Next you feed in additional information – say, about the relative strength of your supply chains. The machine adds in the new information, whirrs again, and the dial flips 180 degrees – based on all available information, you should not launch your product.

But wait – you have more information to feed into the machine. This time it’s focus group data that shows that people prefer your product by a 2-1 margin. You feed this in, the machine whirrs … and the arrow swings back to “launch.”

Next you get cost comparisons. These show that the other guys have a significant cost advantage over you. Feed it in, whirr, arrow swings back to “don’t launch.”

But then your marketing people come in and dump an analysis of price points into the machine. They show that consumers have shown a willingness to pay more for quality – specifically, the kind of quality advantage your product provides. The arrow swivels back to “launch.”

Your sales force comes in and feeds the machine with more information. It appears that there is one chain of stores that is perfect for your product. It has tremendous reach and power and can make your product the undisputed king of this market segment. The arrow quivers with delight and points even more solidly toward “launch.”

So you launch. And your product flops. Why, you ask? Maybe you never find out. But perhaps, after several years, a tongue or two are loosened at some conference, and you find out the truth. It appears that the buyer for the all-powerful store chain has a brother-in-law at your competitor. Feed that into your machine and watch the arrow swing hard to “don’t.”

Now, a machine like this does not exist in the real world, unless you count a magic eight ball, which does not ordinarily accept market research data. Nor are most decisions as dichotomous as the one I posit here; it is never simply “launch/don’t launch,” it’s also “launch with this or that supply chain strategy,” or “don’t launch and sell the product to such-and-such a competitor to get some return on it.” But this abstract approach does allow us to draw some lessons.

The first is that any decision model is completely dependent on the information that goes into it. If you do not have the right information, even the best decision model – or strategy – is likely doomed. From our example, one might conclude that you need as much information as possible. But a second’s thought banishes this conclusion. In theory, anything in the universe could be relevant to your decision. Does that mean you need to input all the information in the universe into your decision model? Maybe you do, logically, but in the real world, you cannot, both for reasons of space and time. By the time you gather all the data in the universe that might be relevant, the opportunity for a decision is likely to be either gone or completely irrelevant.

On the other end of the spectrum, you do not want to simply make decisions based on no data at all. So a strategic decision, by definition, must be made under conditions of uncertainty. A wise strategist, then, must choose what information she is going to input into her decision model – and which information she is not even going to try to get.

And beyond that, a wise strategist will be prepared to be wrong once in a while. That is not the worst thing in the world. Because failure generates data as well. In fact, a certain amount of failure is usually necessary in order to achieve any significant success. Multiple failures, as a matter of fact. The least strategic thing anyone can do is to persist in a failed strategy out of fear of admitting failure. As Edward de Bono once wrote (I am paraphrasing here), if you are digging in the wrong place, you will not find the treasure by digging deeper. You must dig in many places.

This implies, further, that, all other things being equal, you should choose failures that provide good data, and failures that will not kill your company entirely. Choosing manageable failures may be a more important element of your overall strategy than achieving initial success.

As I said, it is not often that you can formulate a decision in a simple binary on/off up/down yes/no dichotomous way. Nature does not too often give us the “either straight north or straight south” scenario. One alternative form a decision might take is continuous. For example, trying to guess the correct value of a certain variable. In this case, the “arrow” could end up pointing anywhere on the 360-degree dial, or, alternatively, anywhere from 0 to x. You have to take your data and try to get as close as possible to that value in order to succeed.

This kind of decision, paradoxically, might actually be easier to handle. It is well-known (at least to those of us who have read The Wisdom of Crowds) that in guessing a single value, the more people you get guessing, the more likely their averaged-out guess is likely to be. If you yourself try to guess how many jelly beans are in a jar, or the weight of a prize steer, you probably have very little chance of getting close. But if you have 100 people guessing, the average of their responses will be a much better guide, normally.

But the first, “yes/no” approach does epitomize much more realistically one damnable aspect of most strategic choice: often, the right choice is one among a handful, and any other choice among that handful is going to be wrong.

Take an investment decision. Half the stock-buying public thinks a certain stock is worth $5; the other half thinks it is worth $25. The price, therefore, is about $15, which in theory may represent some higher wisdom, but in fact, one of the two groups is right, and one is wrong, and there is no plausible case to be made that the stock is actually worth anything in between $5 and $25.

The difference in estimates of the value of the stock between the two groups results from two possible sources: the possible asymmetry in information between the two groups, and the possibly different ways the two groups go about calculating value based on any given information.

The strategic choice here is to go all-in on either $5 or $25. So you go out and try to get as much relevant information as possible. Your first pass has the arrow pointing straight at $5. You get some more information, and the arrow flips over to $25. You keep doing this, and the arrow keeps flipping back and forth, until you satisfy yourself that it is either one or the other, or else until you simply run out of time, and then you have to make a choice. If you are lucky, the information will be predominantly going in one direction, and your level of confidence in your choice will increase. If you are not lucky, you just get more confused, but still have to make a decision.

Strategy is often like that – you are peeling an onion. With each layer, your judgment of the real situation may flip 180 degrees. But eventually, you have to stop peeling, or you will start to cry.