applications of science to the arts of life. - Charles Babbage— 'Future Prospects, On the Economy of Machinery and Manufactures, (First ed 1832)
I used to look at graphs in economics, pricing analysis in marketing, and production functions in operations and think “What a pedantic, academic fantasy. Who has the means to set-up, much less analyze and use such calculations?” My position has changed.
Of course, i was right in one sense. In most cases, most firms don't actually generate the calculations. Reasons might be:
1. The relevant data is not available.
2. If it is, the cost of capturing and analyzing the data isn’t worth the effort.
3. The management team doesn’t have the know-how or inclination to bother.
Most of the time it’s number three—due to an intuitive sense of number two. And most of the time that’s fine. Just understanding that you can and should optimize your product mix helps. Just understanding that minor pricing adjustments can swing purchase volume helps. You often don’t have to be that precise—being directionally right can support the right moves. So, most don’t have to do the actual math.
But, great leaders eventually come to embrace the analysis. After all, it's the next step in what they’re already doing—what they know intuitively works. For example, in the hey-day of manufacturing, leading firms eventually figured out that production functions were indeed helpful. Initially, just being directionally right and understanding that your plant had to be efficient to compete was enough. But as industry became more lucrative, scaled, and thus become more competitive, millions could hinge on small swings in production. This pattern repeats time and again: with industry maturation, methods standardize, competition tightens, and precision matters more and more. In that industrial era, the super-metric became production efficiency. As a result, firms like GM and US Steel hired departments of mathematicians and engineers and used their numerical know-how to outpace rivals. Soon more manufacturers caught on and the exception became the norm. Everyone did the math.
Over the past 15-20 years we've seen the same trend in pricing analysis. Like production functions, hard-core pricing analysis eventually made it’s way from classroom theory into company practice. It’s easy to see why. In the early stages of an industry, you can more or less wing your pricing. But, as competition tightens, you have to be more precise. A few cents too high and you lose the job; a few cents too low and you may win the job but lose money. We’re incented to analyze pricing. Additionally, the marketing function has mushroomed in sophistication, putting powerful tools into the hands of more employees. Analysts and analysis are everywhere. As a result, pricing capability has become increasingly critical, especially in consumer, high volume, and high-ticket sales environments. Here too, we see the math.
So we’ve watched the progression into deeper quantitative analysis (the math) in the production/operations areas and in the marketing areas. What’s next? Without a doubt, it’s HR and performance areas. And if I had to pick the next “production function” or “pricing analysis” I'd wager on learning curves.
A learning curve is a graphical description of a person or group’s learning rate. Setting up a learning curve requires detailed inputs on effort and performance. Thus, like a Profit-and-Loss statement for a savvy manager, the learning curve analysis can provide an enormous amount of insight for a performance analyst.
Why are we concerned with learning rates? Today, the make-or-break is in human capital development. Staffing and performance issues are what keeps leaders up at night. More specifically, Time-to-Proficiency* is as much of a super-metric today as production efficiency was in the manufacturing era. The LC essentially enables deeper analysis of Time-to-Proficiency.
And, today, we have better tools and better data access—so the learning curve on learning curve modeling (so to speak) should be faster. There’s no real excuse not to dive in.
Nevertheless, I suspect we’ll see the same pattern with learning curve usage. Most people in most cases won’t be generating the analysis to use them. That is, they won’t be doing the math.
At least not at first. For the advanced players will catch on soon; and, inevitably, we’ll see fast progression to critical mass. Again, as a given industry becomes more competitive, small differences matter. Today, labor is the vital input. When your firm can do more with given inputs of labor, you’ll command an advantage. This requires more than team-building activities and brainstorming with whiteboards: it requires rigorous analysis.
In short, I’d liken learning curve modeling to pricing analysis or production functions: an esoteric, wonky thing that will eventually become common-place.** The HR and performance management functions will increasingly shift away from the “soft” sciences and into the hard. Once again the math will come into play.