Worst Measurement Mistakes of All Time #73: New Coke

Few product launches went as badly as Coca-Cola’s 1985 debut of New Coke. The carefully researched new soft drink was (allegedly) the first recipe change in Coke’s 100-year-old formula and designed expressly to beat Pepsi in taste tests. Coca-Cola was so confident in their new product that they replaced their old Coke recipe, taking original Coke right off the market. Yet New Coke met such vehement customer resistance that Coca-Cola was forced to reintroduce Coke after just seventy-nine days.

What went wrong?*

Coke’s fading market share spurs a makeover

After World War II, soft drink market share for Coke was 60%. But by 1983, it had declined to under 24%. This was partly because of competition from Pepsi-Cola, but also partly because the overall market for colas steadily declined in the early 1980s. Consumers increasingly preferred diet and non-cola soft drinks, many of which were sold by Coca-Cola itself. Coca-Cola decided to update its formula.

Challenged by the Pepsi Challenge

Remember the Pepsi Challenge? An ingenious advertising gimmick developed by Pepsi had consumers do a blind, side-by-side taste test of Pepsi vs. Coke. Pepsi, distinctly sweeter than Coke, routinely won this battle of the soft-drink sips. There was a critical flaw in this test, however, one that Coke apparently failed to understand (or at least exploit) at the time: In the real world people don’t drink just small sips of soft-drinks, they guzzle cans of the stuff. And when consuming those larger real-world amounts, the less-sweet Coke was often preferred.

New Coke, purposely designed to beat Pepsi in the side-by-side test, turned out to be way too sweet for a lot of people who were going to drink a whole can or bottle of the stuff. This was especially true for diehard fans of old Coke. Still, years after Coke’s exit and prompt reintroduction, New Coke continued to be tested in blind side-by-side tests with New Coke and Pepsi, and preferred by a small percentage of drinkers. Just goes to show, lab research doesn’t always generalize to the real world.


Turns out the color and design of food packaging has a lot to do with how we perceive the flavor of the product inside. Coca-Cola failed to account for this in its taste tests. Many Coke drinkers just could not stomach New Coke’s packaging, which they, literally, found to be less tasty.

No accounting for taste and sentiment

The testing and development of New Coke was carried out under very tight secrecy and security. The new formula appeared to be preferred to both regular Coke and Pepsi in taste tests, surveys, and focus groups. Yet those who tested the potential new versions were not informed that their choice might result in the replacement of the old favorite, one of the most beloved beverages of all time. Thus Coca-Cola under-estimated consumer’s sentimental attachment to Coke.

Upon the launch of New Coke, a very vocal and upset group of old Coke drinkers absolutely refused to drink the new version, and were not at all shy about letting the rest of the world know about it. Many of these drinkers were Southerners, some of whom considered Coca-Cola a fundamental part of their regional identity.

Marketing ego

Coca-Cola was fixated-to-a-fault on beating Pepsi. Coke’s market share had already been cannibalized by it’s own Diet Coke launched in 1982, and then Cherry Coke in 1985. So Coca-Cola didn’t want yet another cola option to dilute the dominance of its flagship beverage. It was so intent on pushing New Coke to be #1—and so confident in the new formula—that it dropped old Coke completely when New Coke launched. This meant old Coke fans had to either go New or go elsewhere.

In hindsight we can also fault contradictions in Coke’s egocentric advertising slogans: if Coke Is It! and The only thing like Coca-Cola is Coca-Cola itself! then what was New Coke? Coca-Cola was narcissistic enough to assume its fans would simply swallow what was obviously no longer The Real Thing.

The return of Coke

New Coke’s reception was initially favorable; sales and Coca-Cola’s stock price went up. But soon the vocal backlash came on strong, sales leveled off, and protests from bottlers and the public meant that people were trying and/or switching to alternatives to New Coke. As a result, Coca-Cola lost several percentage points of the soft drink market. And so Coke was reintroduced, as Coke Classic, after just three-and-a-half months off the shelves.

Still, looking on the can-half-full side, the resurrection of Coke was a huge event, making front page headlines. Two major TV networks interrupted their regular programming to break the news. Three months after the reintroduction, Coke Classic was selling better than both New Coke and Pepsi. New Coke lingered but faded into obscurity. It was renamed Coke II in the 90s, and finally discontinued in 2001.  ∞

* To bring our readers an interesting story we stretch, somewhat, the strict definition of measurement. The development of New Coke and the premature death of Coke is a classic business and marketing research case study, dissected in-depth by thinkers as varied as Malcom Gladwell in Blink (Little Brown, 2005) and contemporary digital marketers. We encourage interested readers to do more research. The brief summary presented here relies mostly on Bill Fawcett’s 100 Mistakes that Changed History (Berkley, 2010).


About Author

Bill Paarlberg

Bill Paarlberg is the Editor of The Measurement Advisor. He has been editing and writing about measurement for over 20 years. He was the development and copy editor for "Measuring the Networked Nonprofit" by Beth Kanter and Katie Paine, winner of the 2013 Terry McAdam Book Award.