The Balanced Scorecard

Beginning today, I am posting all my book reviews to this website as well as my LinkedIn site. This blog will allow others to post their comments -- something the LinkedIn book review site does not allow. If you would like to view my profile on LinkedIn, please click here. If you would like to see my previous book reviews (all 70+ of them) you will need to click the "View Full Profile" button on LinkedIn.

Kaplan and Norton's Balanced Scorecard has become something of an icon in the business literature. Originally published in 1996 it has seen many reprintings. The fact that the title is still available is a major testament to the quality of the work. The basic thesis of the book is this: metrics for Financial, Customer, and Internal Business Processes, along with metrics for Learning and Growth are all vital to corporate growth and must be considered in tandem. If we wish to score financial health through some metric -- say return on net assets employed (RONAE) -- we should not do so by shipping shoddy products early. In other words, there is a right way and a wrong way to achieve financial metrics. The wrong way increases a financial metric at the expense of another dimension of corporate health. Only by scoring all metrics simultaneously, can we get a true picture of the enterprise. I remember the main four metric areas with the acronym FACILE (Financial And Customer, Internal business process, LEarning and growth). Well, it works for me.

Because actual measures are always lagging indicators, we need to construct some hypothesized model using leading inputs for each of the four balanced scorecard dimensions. For example, in the Customer dimension we can measure customer satisfaction by means of a survey. However, we won't be able to get that result until sometime after the sale and we can't understand from that metric why a customer chose not to use our product or service in the first place. Therefore, customer satisfaction is a lagging indicator. One leading indicator might be the waiting time for the customer to talk to a sales representative or cycle time to respond to a customer inquiry. Tacit in this leading indicator is an assumption that faster service means higher customer satisfaction. However, if orders are filled quickly but inaccurately, faster service will ultimately achieve nothing. So, we add order fulfillment accuracy to our metrics. We proceed in this fashion so as to add a mix of leading (input) and lagging (output) metrics. But even this is not enough.

We can make all our customers very satisfied by efficiently delivering product below cost. But profit will suffer horrendously. So, we must consider financial metrics in tandem with customer metrics; that is, we must balance the scorecard. Kaplan and Norton contend (and I agree) that balancing scores across all four dimensions is needed to maintain a profitable company -- one that is growing to satisfy customer needs (and add value to society, by my way of thinking).

But what if our model is wrong? What if we choose the wrong leading indicators? Because we are not omniscient (or even prescient in many cases) we must continually validate our model by seeing if inputs are actually correlating with outputs; and in the case that they correlate, we must assure that they cause. There is a whole science of statistical model building and validation (something I have studied and practice extensively) that can help here, but organizations can puzzle this out if they actually take the time to look at the data. While we're here, I should say that the authors do not use the terms model or validation though they do describe the concepts in different words. I suppose my process control experience is showing here; but we do have a process -- the process of making money by providing value to society -- and we are trying to control it. The validation step fine-tunes the process. Here, and in other writings, Kaplan refers to this concept of continual fine tuning as double-loop learning. The first loop is the proposed input-output model and the second loop is the man-in-the-loop learning activity which reconfigures the model to match reality. Reality includes changing social conditions, new features of the competitive landscape, etc.

If we have arrived at this point, all should be well, presuming that the model is explicit and coded appropriately into balanced scorecard (BSc) metrics; this cannot be taken for granted. Kaplan and Norton are very explicit here: the BSc must be driven from the top down (CEO level) because success requires enterprise-wide buy-in. And this process is a continual one. The authors devote several chapters and an appendix to actually doing this.

In sum, and considering the many management books I have read, this is clearly the best practical guide for management toward achieving sustained profit in the marketplace. It is consistent with the proper role of profit in society and the value-added nature of capitalistic enterprise. I strongly recommend the work -- even after more than a decade, the book continues to deliver.