Uma R. Karmarkar, Baba Shiv, and Brian Knutson wanted to see how we evaluate a potential purchase when the price is the first piece of information we get, as opposed to the last piece of information. They used both fMRI scanning and behavioral tracking to see how study participants responded. Participants were given $40 to spend and then were presented with a number of sample offers. In all cases, the price represented an attractive bargain on the product featured. But one group was given the price first, and the second group was given the price last.
There was another critical difference in the evaluation process as well, meant to help understand the balance of risk and reward in our purchase decisions — a topic I’ve been talking about for years.
In the first phase, participants were shown a variety of products that they would consider low risk/high reward, falling into the second quadrant of the risk/reward matrix I introduced in a column five years ago: "Here, we have our little indulgences; the day-to-day treats that make life worth living… These purchases are all gas and little brake. If you ever make a purchase on impulse, it's almost guaranteed to fall into this part of the behavioral matrix."
The researchers were paying particular attention to two different parts of the brain: the nucleus accumbens and the medial prefrontal cortex. For a layman’s analogy, think of you and a five-year-old child walking down the toy aisle in a department store. The nucleus accumbens is the five-year-old who starts chanting, “I want it. I want it. I want it.” The medial prefrontal cortex is the adult who decides if they’re actually going to buy it.
In the study, the researchers found that the sequence in which these two parts of the brain “lit up” depended on whether or not participants saw the price first. If they saw the product first, the nucleus accumbens started its chant: “I want it.” If they saw the price first, the prefrontal medial cortex kicked into action and started evaluating whether the offer represented a good bargain. In the case of the high-reward products, although the sequence varied, the actual purchase process didn’t. In most cases, participants still ended up making the purchase, whether price was presented first or last.
But things changed when the researchers tried a variety of products that fell into the first quadrant of the risk-reward matrix: low-risk and low-reward. These are the everyday items we have to buy. In the study, they included things like a water filtration pitcher, a pack of AA batteries, a USB drive, and a flashlight. There was nothing here that was likely to get the nucleus accumbens starting to chant.
Now, it should be noted that this follow-up study did not include fMRI scanning, but by tracking purchasing behaviors, we can make some pretty educated guesses as to what’s happening in the respective brains of our participants. Here, presenting prices first resulted in significant increase in actual purchases, outnumbering instances when price was presented last. If price comes first, we can imagine that the prefrontal cortex is indicating that it’s a good bargain on a needed product. But if a relatively boring product is presented first for evaluation to the nucleus accumbens, there’s little to excite the reward center.
An important caveat to this part of the study is that the prices presented represented significant savings. After the simulated purchases, participants were asked to indicate a price they would be willing to pay for the product. When the price was the lead, the prices they named tended to be a little lower, indicating that if you are going to lead with price, especially for quadrant-one products, you’d better make sure you’re offering a true bargain.
If anything, this study provides further proof of the value of knowing a prospect’s mental landscape. What are the risk and reward factors that will be motivating them? Will the media prefrontal cortex or the nucleus accumbens be calling the shots? What priming effects might an early introduction of price introduce into the process?
When I wrote about the risk/reward matrix five years ago, one commenter said: “a simple low-high risk/low-high reward graph is not very useful for driving just in time- and location-based offers, discounts, etc.” I respectfully disagree. While more sophisticated models are certainly possible, I think even a simple 2X2 matrix that helps map out the decision factors that are in play with purchases would be a significant step forward. And this isn’t about driving real-time variations on offers. It’s about understanding the fundamentals of the buyer’s decision process. There’s nothing wrong with simplicity, especially if it drives greater usage.