Coca-Cola's New Coke
In 1985, Coca-Cola had clean data: in blind tests, consumers preferred a sweeter new formula, just as the Pepsi Challenge had been showing. So Coke reformulated, discontinued the original, and launched New Coke in April. The public revolt didn't match the research at all, people weren't upset the new taste was bad, many admitted it was fine. They were furious the original had been taken away. By July, Coca-Cola Classic was back.
For founders and operators, this is a case about what your metrics quietly fail to capture. It sharpens the judgment around any major product or brand change, especially when the data looks unambiguous. The methodological error sitting at the heart of this story is one many teams still make today, and the case is built to make you spot it in your own decisions.
Frequently asked questions
What was the New Coke disaster?
New Coke was a 1985 reformulation in which Coca-Cola replaced its original recipe with a sweeter formula and discontinued the classic version. Blind taste tests had shown consumers preferred the sweeter taste, but the public revolted, not because the new taste was bad but because the original had been taken away.
When did Coca-Cola bring back the original formula?
Coca-Cola brought the original formula back as Coca-Cola Classic in July 1985, just months after launching New Coke in April. The reversal came quickly once the scale of the public backlash became clear.
Why did New Coke fail despite winning taste tests?
New Coke failed because the blind taste tests measured taste preference but missed brand attachment and the emotional cost of removing something people loved. Many customers admitted the new flavor was fine yet were furious it had replaced the original. The data looked unambiguous but quietly failed to capture what actually drove loyalty.
What can founders learn from New Coke?
The lesson is that your metrics can quietly fail to capture what matters, especially before a major product or brand change when the data looks clear. A methodological blind spot, like testing taste without testing the value of what you are taking away, still trips up teams today. CaseBook turns this into a move you apply to your own company, with an AI coach that reads your answer.