Skip to content

“Empirically, my dear Watson.”

  • by

If I ask you about when I should expect water to reach boiling point, you’d probably say 100C or 212F without having to deliberate further. However, what if I told you that water’s boiling point for La Paz residents in Bolivia is 87C (188F)? Despite Indian schoolchildren learning that water would reach boiling point at 100C, those living in Karzok, India find that water would boil at a lower temperature (84C / 184F). When somebody reaches the summit of Mount Everest, water would reach boiling point at 68C or 164F. On the flipside, those living in Jericho, in the middle east would have to wait until water reaches 101C (214F) if they want to make a cup of tea.

Does that mean that the law of the nature is broken?

Those who have studied science would know that the truth isn’t as simple as regurgitating the mantra that water boils at 100C / 212F. Boiling point for liquids is affected by air pressure, which in turn would differ by altitude. La Paz, Bolivia sits at 3,625m, Karzok India is at 4,595m, whereas Jericho lies at -258m below sea level. There are boundary conditions to the empirical generalisation that water would boil at 100C / 212F.

When I joined the Ehrenberg-Bass Institute, I learned about empirical generalisation and how the Institute applies this approach and philosophy on its research and discoveries through multiple sets of data across countries, categories, and time periods. Good empirical generalisations would satisfy the following characteristics, as formulated by Barwise in 1995: Scope / Boundary Conditions — having clear guidelines when the generalisations would hold, and when results would deviate; Precision — the generalisations are reasonably exact to avoid any vague or ambiguous interpretations; Parsimony — the empirical generalisations should be able to be expressed simply, like E=mc2; Relevance — the empirical generalisations need to be useful within the context of the matter; and finally, it needs to provide a Basis for theory-building. For further reading, you can check out the following papers: Uncles and Wright (2004) and Ehrenberg (1995) among other papers discussing the topic.

Having this framework and approach in science helps us not to fall into the trap of being overly reductionistic. If we focus on ‘parsimony’ alone without thinking, the moment we find that water would boil at temperatures other than 100C, we would throw tantrum that this science is broken. Residents in La Paz would passionately argue that water does NOT reach boiling point at 100C. This is the danger of single set of data. If only they would gather all sorts of data from places with differing altitudes, they would most probably come to a greater understanding of thermodynamics.

It’s a similar case with Marketing Science. It’s a relatively new field of science – being pioneered by the likes of Ehrenberg, Goodhardt, and Barwise in less than a century. By academic lineage, I’m proud to say that I am one of his descendants. Having had a career in the industry before I joined academia, I can see how the Laws of Growth could explain the things that confused me. Being part of a team that continues to push the envelope further in this area is exciting. For marketers, it should not be enough just to simply accept that penetration is the way to grow brands without understanding its depth: what it means in the context of the businesss, and learning how it’s holding under different boundary conditions. Our knowledge should not be predominantly derived from tweets, soundbites, and 20-minute online seminars, it should be distilled through thinking, learning, reading, looking through data, and strategising.

In the context of marketing, I can also see that the Laws continue to hold even in the era of digital media, disruptor brands, COVID, and global recessions – from breakfast cereals and gardening tools to B2B relationships, and from prints and TV to social media. These differing conditions help us to refine the empirical generalisations and refine the boundary conditions and their precision – rather than saying that they’re broken.

We continue to refine our knowledge through new discoveries – indeed, isn’t this what science is?

Leave a Reply

Your email address will not be published. Required fields are marked *