I Want, Therefore I Buy

How Desire Shapes Consumer Choice, and How We Measure It

A shopper considers a pair of shoes. Will she take it home?


SUMMARY: Wanting something is different from liking it. By measuring the intensity of desire towards a brand, one can predict its marketplace success more accurately than by measuring how much the brand is liked. We have developed a brand evaluation methodology called Qwantum that recognizes this distinction. We use Qwantum for brand tracking, advertising testing, and for measuring customer satisfaction.


It has to be the most frequently asked question in all consumer research:  "Will anyone buy this?"

Billions of dollars are spent every year on customer satisfaction surveys, net promoter scores, brand equity valuations, segmentation projects, and ad tests. At their core, this is the question they all seek to answer.

Because we happen to live in a competitive economy, the question usually takes the form of "Will anyone buy this and not that?"

The marketing industry has a name for that moment when the customer stands in front of the supermarket shelf and makes a decision — the moment of truth. 

Among other things that have been called "the moment of truth": the budget deficit and the global warming.



People buy what they want

One common criticism of advertising is that it makes people buy things they don't want. 

If only.   The road to brand hell is paved with good intentions of companies with big ad budgets that tried to sell stuff to people who wanted none of it.  "Failed to find a product-market fit" is how they write it on the tombstones.

People buy what they want. This truth is so self-evident it is almost tautological.

And yet, we don't ask what people want often enough. We ask about goals, tasks, preferences, values, beliefs, and habits. We measure emotions. It must be the Puritan roots because even theLovemarks jingle the chastity belt of trust and respect.

Desire, the pure blinding lust, is the strongest and the oldest force in the universe, the engine that drives the evolution, the spark that ignites wars.  We are wired for lust, and yet we know so little about it.


I remember when the second iPad came out, back in 2011. It was thin, light, fast, and gorgeous. The first day it hit the stores, I woke up at 5am and walked to the Cambridge Galleria mall to be the first in line.  I wasn't.

They didn't have my model in stock on that day. I came the next morning. And the morning after.

What were my beliefs and values that filled my heart with hope and despair?  What goals and preferences conspired to keep me coming back? 

I don't know. I really, really wanted that iPad.


What is it like to want something?

In our group, we spent a lot of time thinking about what it means to want something, how we can measure it, and how we can use it in our work.

Let's go on a shopping trip.

You are a single guy at a Target store on a mission to pick up a new shower curtain. How much do you really care about shower curtains? Not at all, probably, but you are "adulting" and it’s about time you got one.

You walk over to the bath aisle, and see four options.

Four shower curtain options. Which one would  you  choose?

Four shower curtain options. Which one would you choose?

As you look at the curtains, your brain revs up its weighting machine.

The first one, A, you rule out right away because you hate the color. The next one, B, is not so bad, and it costs the same. Then you see the one with the cityscape, C, and you kind of like that one. The last one, D, is pretty plain, but it’s 10 bucks cheaper than the other three. You have arrived at a ranking: you don't care a whole lot either way, but all things considered, you like E the most, then D and B. Option A is so far down the list that you would rather not have a curtain at all than buy a purple one.

This dimension of consumer choice is the focus of most consumer research. A lot of exceptionally interesting work has been done to understand how people evaluate their options, how cognitive biases influence this evaluation, and how the rankings are constructed.  And lot of marketing dollars are spent on nudging brands to the right of their competition.

Some researchers refer to this as a rational dimension — as opposed to emotional — but this distinction is not accurate.  Ultimately, all of our choices are guided by how we think they will make us feel.  

But regardless, this mental pros-and-cons accounting is only one side of consumer choice. There’s a second one.


Just as you are about to grab curtain E, you turn around and you see this: a shower curtain that looks like a YouTube page with a window in it.

A YouTube shower curtain: it makes my heart sing.

Now, a number of things are happening in your brain at once.
You can already see the party you'll throw next week and the guests lining up to take selfies with your curtain. You are counting all the Facebook and Instagram likes you'll get. You can hear everyone saying how awesome you are.

Psychologists call this “incentive salience”. It means that, first, a cue – in this case, a shower curtain — triggers a desire for a particular psychological reward that is now brought to the foreground and magnified. The cue itself becomes associated with the reward and becomes an object of desire itself. 

You are becoming excited. Your pupils dilate, your heartbeat rate goes up, and your skin's conductance increases. You want to win your friends' admiration, and this YouTube curtain is just the ticket.

If we were to graph how you feel about the curtain situation now, it would look like this:

The YouTube curtain is in the category of one.


Notice how the YouTube curtain may not be the best option based on the selection criteria you were applying to the other four such as the price.

We see it happen all the time. Options that get people excited are often inferior to the available alternatives in some objectively measured way. Fewer features. Smaller size. Higher price. But it doesn't matter anyway. These products exist in a different dimension.

Try telling a teenager that his or her crush is objectively, by any measure that should matter, inferior to other alternatives in the dating pool. When we want something, we want it. Or, as Blaise Pascal put it:

"The heart has its reasons, which reason does not know."

Liking ≠ Wanting

There's a difference between liking something and wanting it. The distinction is grounded in how are brains are wired:

“For positive emotional states, core processes of "liking" and "wanting" are psychologically dissociable from each other. "Liking" corresponds to a basic sensory pleasure or hedonic activation. "Wanting" corresponds to a different core process, the attribution of incentive salience to stimuli or events. Core processes of "liking" and "wanting" are mediated by different neural systems in the brain.” (Berridge, 1999)
Liking and wanting “can be manipulated and measured separately. Liking and wanting have separable neural substrates. Mediation of liking […] involves neurotransmitter systems such as opioid and GABA/benzodiazepine systems. Mediation of wanting […] involves mesotelencephalic dopamine systems. “ (Berridge, 1996)

We believe the two serve different evolutionary purposes. Liking is a feedback mechanism that doles out rewards and punishments to guide our choices.

Wanting, on the other hand,  plays a motivating role. It points us towards goals and it drives us to engage in behaviors that promote individual's survival or reproductive success.

While we typically like the things we want, the reverse is not always true. In other words, we don't always want the things we like, as in "I enjoy reading, but not right now."


Measuring Desire

One common way brands, products, or features get evaluated is by asking people a question along the lines of "On a scale of 1 to 10, please tell us how much you like or dislike the product."

If the options are comparable (so, not apples to wrenches) and meaningfully differentiated from one another in consumers' minds, this approach works fine in predicting consumer choice. Broadly speaking, options that have a higher "like" score tend to get chosen by more people. 

The problem is that for a great number of products, people's "like" ratings will not give you a clear view of the choice they would make in a shopping situation.  One reason is that at least in the U.S., consumers seem to be cautiously optimistic — or perhaps just polite — about unfamiliar but mainstream products. In the data we have collected, "like" ratings tend to cluster in the positive territory at around 6 on a 9-point scale. 

- Do you like this health insurance brand?
- I guess
- How about this one?
- Looks good
- And this one?
- Sure

Instead of asking just the like-dislike question, we ask people whether something excites them, and to what degree.

We plot the results on a grid that looks like this, with pleasure (like-dislike) on the x-axis, and with excitement (arousal) on the y-axis:

This grid is built on two 7-point scales, but it works the same way with scales that have 5 or 9 points.


The names for the quadrants are based on our experimental results: brands in the Want quadrant are 2.6 times more likely to be chosen over competitors than brands in the Like quadrant (r^2 values of 0.74 and 0.28, respectively).

In other words, if we look at where each option sits on the grid, we can can tell with a great deal of accuracy which option will end up chosen.

Speaking technically, the combination of relative positive valence and high arousal is indicative of consumer desire and is highly predictive of consumer choice.

We call this methodology Qwantum.


The Desire Gap

One way of putting Qwantum to work is by measuring how different options available on the market compare to an idealized vision of the product or experience consumers have in their heads.

Close your eyes and imagine a car of your dreams. Imagine how it drives. How it smells. The sound it makes. Imagine what it feels like to have the car in your driveway, and what it feels like to drive it on an open road. Pull this car over by a restaurant, and imagine how that feels, too.

If we ask you to grade whether your dream car makes you feel calm or agitated, and whether it makes you feel happy or sad, and then chart the results, the graph would look like this:

This is your ideal car.

Now, we if we ask you how you feel about BMW,  Audi, Chevy, Ford, or some other brand, and charted those scores as well, we would get a graph like this:

How other cars measure up to the ideal.

How other cars measure up to the ideal.

This distance between a brand and the ideal car is what we call the Desire Gap. The shorter the distance between a brand and the ideal, the more likely you are to buy this brand and not others. 

The Mindshare Monopoly

We also found that while many brands can live in the Like quadrant, there's usually only one brand, if at all, that sits in the Want territory.

Liking is plural. Wanting is singular. Making products people want and not merely like gives you not just a competitive advantage. It gives you a monopoly.

The iPhone is in the category of one. For years, it's been lagging behind the competition on the number of features and other technical dimensions. Yet, many people who buy iPhones don’t go out to shop for a smartphone. They go out to buy an iPhone. Their brain does not instruct them to buy iPhone based on a weighting of pros and cons.  They want an iPhone.  And iPhone has a monopoly on iPhones.

Here are a couple of graphs from the same car research project. The first one shows how people who are hot on Chevy feel about other brands — the only want Chevy.  The second one shows people who are hot on Ford.

People who would rather push a Chevy.

People who would rather push a Ford.


America's Most Wanted

Using a different visualization method allows us not only to rank brands by their desirability, but also to understand in detail how people feel towards each brand.

In another experiment, we used Qwantum to evaluate five casual dining brands, and mapped the distribution of responses by quadrants for men and for women. We then compared the share of respondents in the Want quadrant for each brand.

Men are from Hooters, women are from Olive Garden. (Click to make bigger)

Among the five brands we tested, Hooters, a restaurant chain known for wings and the skimpy orange outfits of its female servers, appears to be the most polarizing judging by the stark differences in score distributions between men and women.   Hooter’s was also the only brand in our experiment that was “wanted” more often than it was chosen in a gift card selection task, perhaps reflecting its position of a “forbidden fruit” in our culture.



Here are some other ways in which we've used Qwantum to help brands:

Market segmentation.  Should the brand focus its marketing efforts on the people who already want it, or should it instead excite people who merely like it?

Brand tracking. Has the number of people in the Want quadrant changed as a result of advertising activity? Do more people Hate the brand after a recent media debacle?

Customer satisfaction. Few brands are truly hated, and few people have brands they truly hate.  For an average brand, the most dangerous area to be in is the Apathy zone.

Advertising pre-testing. How do people move between the zones as a result of seeing the ad? Do more people Want the brand?


How to Make Your Brand Desired

There’s this famous saying about how people don’t buy drills, they buy holes. Only they don’t really buy the holes either. What they really buy is some peace and quiet at home, or a warm memory that comes back every time the look at the picture on the wall, or the admiration of their friends, or a sense of accomplishment.

We call all these things emotional rewards.

There’s a large body of academic thinking that goes back a hundred years about what these emotional rewards are. One of the first attempts to classify them at the end of the 19th century produced a list of 4000 of these rewards – or instincts, as they were called then.

We’ve distilled these rewards down to ten.

But that's a story for another time.

With Blair Ballard, Henry Bruce, Bill Letson, Rob St. Louis, Andres Hernandez, Matthew McKenna, and Justin Holloway.

Special thanks to Lev Mazin at AYTM.