Does Satisfaction Guarantee Loyalty?

I noticed an interesting statistic today, that 80% of "new" customers reported that they were "satisfied" or "very satisfied" with the seller or brand that they had just left. This flies in the face of conventional wisdom, that a satisfied customer is a loyal customer - after all, if their experience was satisfactory, they should have no reason to switch. The eighty-percent statistic also seemed hyperbolic, so I did a bit more research.

From what I was able to find in other sources, the statistic might be exaggerated, but probably not by much. I found a dozen or so references, and most of them were in the range of 60% to 80%, though they defined "satisfaction" in various ways, so I doubt that there can be an agreement on a precise number ... but all the figures I saw were very high.

There is a fundamental difference between the two, which seems entirely rational: satisfaction pertains to the attitudes about a past behavior, whereas loyalty pertains to the attitudes that will influence a future behavior.

And so, a customer who reports being "satisfied" is reflecting on a choice he made in the past and is being asked, in effect, if that choice was an error. This might be a part of the reason that ratings and reviews are overwhelmingly positive, and tend to be egocentric (see my previous note) - as such, an individual who is asked whether they are satisfied is likely considering how the rating reflects on their judment, not their actual satisfaction.

Granted, this is highly speculative, and I don't expect one could develop a survey question that would avoid that bias to separate "real" satisfaction from alleged satisfaction. But even if ego were taken out of the equation, it remains likely that customers would remain generally positive about past decisions, as it's a reflection of known facts: they can be certain that the product they purchased did (or did not) satisfy their need and that the experience of dealing with the vendor was pleasant or unpleasant. It is known.

Loyalty, meanwhile, depends on the unknown, and requires greater speculation about possible future conditions. When a similar need arises in the future, the customer may have doubts that the present supplier or product will meet their needs in the same way. Or perhaps the nature of their needs will be different due to the situation in which they will be when the need arises. Or perhaps they will become aware of additional alternatives they might have preferred had they been available in the past.

Since loyalty deals with future conditions that are not known, it is unlikely that it can be measured, except by monitoring purchasing behavior - and even then, it is only past behavior that can be observed. You cannot assess whether a customer will be loyal, only that they have been loyal thus far, following a pattern that can change at any time.

But neither can satisfaction accurately be gauged by a survey question. It will always be influenced by the conditions under which the buying decision was made in the past, and it is highly likely to be skewed by the desire to escape the embarrassment of admitting a mistake.

And so, to say that a customer "reports being satisfied" with a past purchase means little to nothing. The (objective) fact that the same person has purchased steadily is more telling of their level of satisfaction - and at the same time, past behavior cannot be taken as a a guarantee of future behavior.

In the end, I don't think that customer satisfaction can be dismissed altogether. A person who has a rotten experience is probably less likely to repurchase from the same vendor, and a person who's had an excellent experience is probably more likely to purchase. So there is some connection, though not as strong as many seem to assume.

My sense is that each purchase of a product presents a unique buying opportunity and a unique buying decision, and the notion that "this brand has satisfied my needs in the past" is one of many factors that drive their decision at each instance in which they find themselves in similar circumstances.

The nature of these factors merits more consideration at a later time - but for the present topic, I'm left with the sense that measurements of satisfaction are neither as reliable nor as accurate as they are purported to be, and merit a bit less consideration than I've given them in the past.