Eureka Moments!

by Richard Goosey | Aug 17, 2018

Being a marketing science guru and statistician, I've always been fascinated by better understanding human behavior and the decisions we make that have profound impact on our lives. These are the "eureka moments" of suddenly understanding a previously incomprehensible problem or concept.

Interestingly, even though they are sometimes very profound, they can at times be discounted or ignored. For example, in an eye-opening Reuters article published in June 2014*, it was revealed that “Strong Atlantic hurricanes with the most feminine names caused an estimated five times more deaths than those with the most masculine names”. Why? Because people tend to believe that a “feminine” hurricane will be less destructive. Behavioral scientist Hazel Markus of Stanford University, who was not involved in the study, called it "proof positive" that gender stereotypes affect behavior even when the male or female isn't a living thing. Understanding that the public is going to be less concerned and behave differently when a hurricane called Samantha is heading their way is a profound eureka moment as action to counter this behavior would greatly reduce the number of deaths in the future!

However, we don’t always act in ways that are clearly logical. We don't need to look any further than the excessive time taken to move from association to causation between smoking and lung cancer. While the first study was carried out in 1950 by Doll & Hill based on real world research with 53,000 doctors, and the initial results published in 1954 in the British Medical Journal, the government played down the research – with the Minister of Health chain smoking throughout the press conference and stating that “the evidence is statistical only”! Even with the Doll & Hill study revealing by 1960 that the death rate from lung cancer was 13 times the rate in smokers than non-smokers, and 30 times in heavy smokers, it wasn’t until 1965 that TV advertising was banned in the UK.

Experiencing an eureka moment, where we establish a leading indicator of future events, is "gold dust" in marketing science. And I believe we've found a leading indicator for future pharmaceutical brand performance by measuring doctor drug preferences versus their prescribing behavior. The leading indicator being if a doctor’s share of preference for a medicine is different than what they actually prescribe, doctors are likely to change their behavior in the future to align with their underlying preference – which clearly impacts brand performance.

While clearly not rocket science, isn’t this what brand teams work to achieve? With brand teams' marketing activities and spend aimed at generating preference for their brand, a key tool for managing a brand's future direction and securing a clear view of the likely ROI of future actions is the ability to measure and track how preference changes against prescribing behavior.

In Kantar Health's latest Edge of Insight report "Marketing Science Innovation: The Road to Weatherproofing Your Brand", we highlight case study data in a landmark piece of research that demonstrates that this strong leading indicator enables impactful decisions based on evidence. This includes leveraging the Zipf distribution, which captures the non-linear way in which people disproportionately value “winning” brands, to transform share of preference to future market share.

Case Study: Market Share vs. Preference Share

Taking three different brands, the following three charts show how they performed over seven quarterly tracking waves in terms of market share and preference share.


What the Data Shows
Preference share is higher than market share, illustrating an opportunity for the brand team to realize the equity they had built in the marketplace and increase their market share. From time periods Q1 to Q4 they managed to achieve that, but after Q4 preference share started to decline and, in tandem with it, so did market share.

Key Learning
This presents a very clear example of how a forward-looking indicator can show a brand team what they could achieve and what to do once they have achieved a higher market share.


What the Data Shows
When market share was higher than preference share, there was an opportunity for the brand team to defend the market share they had achieved – not doing so sees it decline, because preference share was lower than market share.

Key Learning
Here, the model functioned as an early warning system for the brand team, showing them both the need to defend their market share as well as the consequences of choosing not to.


What the Data Shows
An urgent need for course correction as preference share and market share both decline. That begins at Q4, where preference share starts to increase, and market share follows, again proving the former to be a leading indicator of the latter.

Key Learning
Course corrections can be achieved - in the case of this brand a piece of market access material was successfully activated by the brand team to bring about the change shown.

A Eureka Moment in Marketing Science

The revolutionary new uses of the “share of preference” measurement is a eureka moment in marketing science for the pharmaceutical industry. Combined with a brand’s existing market share, share of preference can now be used to look at where a brand’s market share currently stands, where it could rise to, and most importantly how to get there by pulling the lever and minimizing the impact of barriers!

Read more in our Edge of Insight report "Marketing Science Innovation: The Road to Weatherproofing Your Brand".

*Source: "Female Hurricanes are Deadlier Than Male Hurricanes" (University of Illinois at Urbana-Champaign - Proceedings of the National Academy of Sciences).


Leave a comment

More than one Google Analytics scripts are registered. Please verify your pages and templates.