There’s an awful lot of talk during this coronavirus pandemic that we need to base decisions on the science. The “science” is informed by data and vice versa. But how good and, therefore, useful is the data? After all, we must be mindful of GIGO (not GEICO, but GIGO), the acronym for “garbage in, garbage out.”
What are we to believe when New York reports a lower rate (%) of deaths in nursing homes from COVID-19 than the rest of the nation and other hard-hit countries? How reliable is the data from NY when they are not ascribing the deaths to nursing homes if the patient was transferred to a hospital and died there? How accurate is data on the number of deaths from COVID-19 when the government compensates hospitals for each patient who succumbs to what the hospital chooses to classify as COVID-19, and even more when put on a ventilator? Might some of these patients died due to other causes such as age or life-threatening comorbidities? Is the data corrupted by “gaming” a term used and expounded by Jerry Z. Muller, in his book, THE TYRANNY OF METRICS? (The book was gifted to me by my friend, Bill Weintraub, the former CMO of the Coors Brewing Company? I found it enlightening yet frightening—especially during these difficult times.)
The world of business continues its rush to BIG data. Data is married to science. Galileo, hailed as the father of science, applied mathematics to observable data. However, science can be corrupted when tied to compensation for specified results (outcomes), by biases, and, yes, by personal agendas. Galileo experienced bias against his science when the Roman Inquisition tried and found him “vehemently suspect of heresy” and sentenced him to house arrest for the remainder of his life. Keep in mind the adage, “Statistics don’t lie (or do they?), but liars use statistics to lie.”
If one doesn’t follow the data, one is accused of being ignorant of science. But from what data is the science drawn? It’s possible that the data is ignorant (drawn improperly), or the manager using the data is ignorant. That’s a topic for next time.
What should we believe and use to guide our decisions when commanded to stay in and shelter when approximately 50% of deaths (if this is a reliable data point) occur to nursing home patients who stay in and sheltered (or not, as sheltering suggests protection)?
In the 16 May 2020 edition of the WSJ, Allysia Finley published her interview with Aaron Ginn, a Silicon Valley technologist, in “The Lockdown Skeptic They Couldn’t Silence.” Mr. Ginn questions the scientific validity regarding our 6-foot rule for social distancing, among other data about COVID-19. He points out that the World Health Organization “recommends 1 meter (3 feet, 3 inches), while Germany and Australia suggest 1.5 meters (just under 5 feet).” Which one should we believe, if any? Is it science because it is data? Does each group march to a different science? Should we believe any of the numbers—or, even, what we read? Perhaps, we should be skeptical of the data and what purports to be science. At a minimum, we need to dig deeper.
Now, you may think this musing is political; however, it is not. I’ve not taken or presented a stand. I’m merely using it to question data, its tie to science, and whether we should accept it blindly. This is rather practical as it has implications for our marketing and business. Case in point: the introduction of New Coke. It launched in April 1985, most likely before your time in marketing and, perhaps, on this earth. Regardless, you are undoubtedly aware that it was one of the biggest marketing blunders of all time. How’s that? After all, the marketing research supported it.
The marketing research data showed more Coca-Cola drinkers preferred the taste of New Coke to what is now Classic Coke. But the data was gathered via a sip (not extended use) test and sans a concept to inform respondents that the formula for brand Coca-Cola was changed. Therefore, how reliable is the data?
If you want a realistic perspective on taste preference, one must go beyond sip testing and investigate extended use testing. Many of the exclusive Coke drinkers consume significant quantities of the soft drink. So, extended use testing is critical. Also, if you are going to announce your change, then make sure that your customers are comfortable with going about changing something they hold dear—the “brand,” not the product nor the formulation!
Lastly, the marketing research bunched Coke drinkers from three usage segments: those who Drink Coca-Cola Exclusively, Drink Coca-Cola Most Often, and Drink Coca-Cola and Pepsi-Cola Equally. While the composite data was favorable for New Coke, the results among Coke Exclusive and Most Often Drinkers were damning. (By the way, don’t think you are going to bring in more users than you alienate with a 100-year old brand.) You don’t ever want to alienate your most ardent customers, who make up the bulk of your business.
The data was generated scientifically. However, it was bad science. The resultant data was faulty, which was borne out by the market’s reaction to New Coke. It bombed!
Data can be sound or faulty, which makes the science equally so, and vice versa.
So, how good is your data? You need to know as it will impact how good your marketing is!
Next time, I’ll explore what does your data mean.
Take your marketing to the next level. Check-out my new book, AVOID CRITICAL MARKETING ERRORS: How to Go from Dumb to Smart Marketing. It can help you achieve success during and post the COVID-19 recession. Learn more here: http://bdn-intl.com/avoiding-critical-marketing-errors
Stay safe and be well.
Peace and best wishes,