Coke versus Pepsi

Coca Cola (Coke) enjoyed a significant market share lead over Pepsi Cola (Pepsi) in the early 1970's. Eighteen percent of soft drink users claimed to be exclusively Coke drinkers versus 4% exclusively Pepsi drinkers. However, by the early 1980's, the percentages changed to 12% Coke vs. 11% Pepsi. Pepsi had been running the "Pepsi Challenge" taste test commercials for many years and, it seems, with great success. Coke executives were distraught over the shrinking market share and, not wanting to believe more people preferred Pepsi's taste, conducted its own taste tests. Coke achieved exactly the same results as Pepsi found: 57% of those tested preferred the taste of Pepsi over Coke (a 16% difference). Testers referred to Coke's taste as "harsh" yet Pepsi's as "smooth" or "rounded." So, Coke went to work and eventually created a sweeter, more "Pepsi-like" tasting "New Coke" that actually beat Pepsi in taste tests. In announcing the move to New Coke, Coca Cola's then-president called changing solely to the new beverage "the surest decision this company has ever made." But within four months, consumers virtually quit buying New Coke altogether making this possibly the biggest marketing blunder in American history. Thinking about the value of research and interpreting research results, how could this disaster possibly have happened? (Hint: Pepsi did not gain additional market share from Coke's marketing blunder.)

Note: The answer should be about how Coke used the research it did and came up with a wrong conclusion.