Richard Oliver’s 1980s cognitive model of consumer expectations further explored the idea that expectations create a reference frame for judging satisfaction or dissatisfaction. And there is, as you’d expect, agreement among most researchers that discomfirmation (plus or minus) does influence post-purchase evaluation—that seems intuitively obvious.

However, there are a number of limitations to the expectancy-disconfirmation model, particularly when it is applied to tourism. For example, everyone seems to know what satisfaction is, but it obviously means different things to different people—researchers and especially consumers. And Oliver agrees with that, suggesting consumer’s minds are a black box; we can see what goes in and what comes out, he said, but we can’t know what really goes on inside.

Worse, customers present marketers with an astounding collection of biases, as we’ve seen. Perhaps one of the strangest is resistance to disconfirmation.

Festinger wrote a fascinating book titled When Prophecy Fails about apocalyptic end-of-world cultists who believe they would be saved by aliens when the rest of world ended. Festinger joined the group to see how they would respond when, he presumed, the world didn’t end. The group’s leader neatly solved the problem, conveniently saving the group from painful cognitive dissonance, by proclaiming that God had communicated (via automatic writing) that Earth would be spared because the group was so faithful.

The thing is, kooks aren’t the only ones who suffer from disconfirmation bias. Asked if he was worried that empirical tests would disconfirm his theory of relativity, Einstein famously said “Then I would feel sorry for the good Lord. The theory is correct!”

Economist John Kenneth Galbraith understood the problem, too, and wrote: ‘Faced with the choice between changing one’s mind and proving that there is no need to do so, almost everyone gets busy on the proof.‘

So why do people resist admitting they’re wrong? Yale psychologist Dan Kahan’s work is illuminating.

The urge to maintain status within one’s social network is so powerful, …., that well-educated people will use their information-gathering and computational skills to marshal a more impressive body of evidence in support of whatever identity it is (freethinking skeptic, caring mother hen) that earns them brownie points in their troop.

For marketers, disconfirmation bias makes the seemingly obvious relationship between expectations, performance, and satisfaction a lot less clear. For that matter, Oliver uses an overall “better than expected – worse than expected” scale and actual perceived performance isn’t considered as a measurable variable. A method for including a scale of perceived performance and a measure of disconfirmation bias into predictions of post-exposure consumer intentions and choice behavior seems an avenue for useful future research.

Indeed, given the complexity of the very notion of satisfaction, it makes the idea that expectancy-disconfirmation model can serve as a universal solution questionable. Perhaps a more accurate picture of customer satisfaction could be obtained by melding the strengths of Equity Theory, Value-Percept Theory, Comparison Level Theory,  and the Expectancy-Disconfirmation Model?