Many years ago, some terrific academic research found that when you plot the logarithm of unit sales produced against the logarithm of unit cost, the result in many manufacturing industries was a straight sloping downward line. This insight was taken to heart by the Boston Consulting Group who developed the famous growth/share matrix.

The reasoning went that if you could gain large market share in a growth market, you could capture a major cost advantage. That would give you a competitive advantage over smaller-share rivals.

You remember the matrix, of course: the 2x2 grid in which you plot the growth rate of your market against your position in that market. The high/high box (big shares in growing markets) were “stars”; the high/low box (big shares in slow-growth markets) were “cows”; the low/low box (small share in slow-growth markets) were “dogs” and the remaining quadrant (small share in low-growth markets) were question marks.

The strategy advice was to invest in stars, use the cows for cash, sell off the dogs, and … well, it was never quite clear what to do with the question marks. At one time, an academic study found that 75% of all CEOs of American companies were aware of and had used some aspect of the BCG matrix in making portfolio allocation decisions. It later transformed into the famous GE Matrix and also found its way into other tools offered by consulting firms such as McKinsey.

Well, it was too good to last, I suppose, because as the model gained in popularity, criticism of it grew. Observers argued that it was fundamentally flawed and led to starved cows, mis-fired stars, lost opportunities for profit and worst of all, the wholesale abandonment of markets whose domestic growth might have stalled, but which were growing globally (such as televisions).

The academics weighed in as well, with studies by Columbia’s own Don Hambrick and Ian MacMillan empirically testing the conclusions in the model (see references). But wait — it’s back!

In the May issue of the Harvard Business Review is an article (“Is Your Growth Strategy Flying Blind?”) on growth strategies that advocates a granular approach to analyzing possible markets, based on — you guessed it — market growth rate and market share (among other things). The approach differs from the old BCG approach in that the units of analysis the authors suggest are smaller — right down to individual product lines, customer segments and regions, but the strategic advice remains pretty much the same. Invest in those segments that show high growth rates, in which the position is strong and in which there is momentum. A great idea whose time came … and went … and has come again.

References: Hambrick, D. C., I. C. MacMillan & Day, D. L. 1982. Strategic attributes and performance in the BCG matrix: A PIMS based analysis of industrial product businesses. Academy of Management Journal, 25(3): 510-531.

Hambrick, D. C. 1982. The Product Portfolio and Man's Best Friend. California Management Review (pre-1986), 25(000001): 84.