Freshplum Blog

The latest in
data science and
online retail from
San Francisco

How to raise prices without pissing your customers off

No matter what size a retailer is, there’s no better way to annoy your customers than by raising your prices. Sometimes, however, you may find yourself needing (or wanting) to do just that. Figuring out how to tactfully increase your prices is tricky, but behavioral economics can help. The key point to emphasize is that consumers tend to be very sensitive to how fair they think a price is, and targeting that experience of a price being fair or unfair can help alleviate your customers’ frustration.

16-tagPrice Fairness

The idea of “price fairness” is a bit vague, but in general, it’s the relatable feeling that a price is about right for a product. When you walk past (or scroll past) a product and you immediately get the sense that it’s too expensive or that it’s surprisingly cheap, you’ve made a price fairness judgment. We usually think of this as an emotional response to price, and as a result, it is sensitive to customers’ existing emotions: customers in a good mood are more likely to think prices are fair.

8-zappossalePreviously, I have discussed how to use a certain type of price fairness evaluation, namely the use of reference prices, to make deals look better to customers. This approach plays on the emotion of price fairness by giving customers something to base their gut-reactions on in order to get an immediate sense that “this is a good price!” or “this is not a good price!”

While this strategy can allow you to temporarily increase customers’ valuations of your products, what if you need to actually increase your prices? How do we address the negative emotion of unfairness?

A comprehensive look at price fairness perceptions comes courtesy of a 1986 paper by behavioral economic heavyweights Daniel Kahneman, Jack Knetsch, and Richard Thaler. The authors conducted a massive set of telephone surveys asking respondents about hypothetical situations dealing with issues like price increases or wage decreases in domains from groceries to photocopies to real estate. They had these people evaluate how fair or how acceptable these company behaviors would be, and one of their most surprising findings was that consumers were, on average, surprisingly sympathetic.

Dual Entitlement

Remember how consumers think prices are more fair when they compare favorably to a reference price? It turns out that 75% of consumers feel that retailers are also entitled to similarly “fair” levels of profit. In other words, if a company was making $X of profit and their costs went up, consumers agree that it is fair for prices to increase so that the company could maintain its $X profit.

In fact, most consumers believed it would be fair for a firm experiencing losses to shift that loss completely onto their customers. As long as the firm was upfront and honest about why it had to raise prices (and those reasons were good), consumers indicated that price increases were entirely fair. Note that this does not necessarily mean they would still be willing to pay higher prices, but it does mean they would not develop a negative impression of the firm.

There are a few caveats to this idea of reference profit entitlement:

  • Just because consumers agree that retailers are entitled to a reference profit does not mean they will agree on what that reference should be. If a firm is already making more money than a close competitor, consumers may instead select the competition’s profits as the most appropriate reference. In a sense, they are making a fairness judgment about how to make another fairness judgment, and successful and profitable companies will generally be seen as having less of an excuse to raise prices (consumers see profitable companies as about 3 times less justified in raising prices as compared to companies losing money).
  • It is not fair to increase prices when the threat to profit is unrelated to the product at hand. For example, increased manufacturing costs may be a fair reason to raise prices on a product, but increased employee pay may be a less fair reason because it feels less directly related. If you want to increase the price of a product, emphasize that it is because of costs specific to that product.
  • It is not fair for a firm to increase prices in anticipation of lost profit. For example, if you know that the wholesale cost of a product will increase at some point in the near future, 79% of consumers would say it is not justifiable to increase the price of your products in stock now. A company must be in a position where its profits are actively threatened in order to fairly raise their prices.

Interestingly, while consumers feel that it is fair for a firm to raise prices when their profits are threatened, they do not tend to think it is unfair if a firm does not pass on savings to their customers. In other words, if costs go up, then retailers are allowed to raise prices, but if costs go down, retailers are not expected to lower prices. Consumers are apparently much more retailer-friendly than we might have thought.


To quote the paper, “the cardinal rule of fair behavior is surely that one person should not achieve a gain by simply imposing an equivalent loss on another.” Consumers are much more forgiving of a price increase when they understand that there was a need for it, and explaining that need will generate more sympathy than ignoring the issue or trying to offer a small concession in conjunction with the price increase.

16-shovelThis also means that retailers must ensure that they never appear to be abusing their position within the market. In the most famous hypothetical from this paper, respondents are asked to imagine that there has just been a large snowstorm in their town, and in response, a hardware store increases the price of snow shovels. From an economic point of view, this is rational: consumers value snow shovels more, so the store is wise to raise the price. However, 82% of consumers see this as an abuse of market power, and even if they need and can afford the shovel, they are both less likely to buy it and certainly more likely to remember this unfair behavior in the future.

Nevertheless, overcoming this bias using explanation can be surprisingly effective. For example, Uber‘s surge pricing (in which multipliers are added to the price of a ride during periods of high demand) is a textbook example of an “unfair” pricing strategy. However, the brief explanation offered by the app that this multiplier is meant to “get more Ubers on the road” manages to diffuse much of the frustration. Lyft goes one step further and states explicitly that the extra fare goes directly to the driver, thus linking the need for the price increase with its source.

To make a price increase seem even more fair, you can mix the explanation strategy with a reference price strategy. For example, when Amazon Prime recently increased it’s annual cost from $79 to $99, not only did they attempt to explain that it was to cover rising fuel and transportation costs, but they also originally suggested that they might raise the price to $119, which makes the $99 price tag look more attractive.

What might be an example of a bad explanation? Just yesterday, Netflix announced that they would be raising prices, and their explanation was, ”these changes will enable us to acquire more content and deliver an even better streaming experience.” This explanation is vague and does not appeal to any underlying costs or losses that a price increase will address (even if there is one). Since it doesn’t do much to address the sense of unfairness that new customers will be feeling as they pay the higher prices, it will be interesting to see how consumers react once the price hikes take effect.


  1. Customers may be surprisingly understanding of the need to increase prices. They tend to believe that firms are entitled to their profits and are fully justified in protecting those profits.
  2. Explanations go a long way. Making your reasons for price increases clear, simple, and fair will go a long way towards making your customers accept your new prices.

Guest post by Stanford’s Alex DePaoli

Follow him @AlexDePaoli.


The Fluency Effect

When designing a good website or a good shopping experience, simple is almost always better. Intuitively, an uncluttered layout with easy-to-follow instructions removes a lot of the friction that customers might otherwise encounter when interacting with an  online store (or any kind of website or mobile app). This makes it easier and thus more likely to be used, perhaps helping to explain the current popularity of minimalist website design.

15-minimalistThe power of simplicity has been further explored by consumer psychologists and behavioral economists. The more simple a tool, whether analog or online, the more easily people feel that they understand it. This feeling of “ease of understanding,” sometimes called Processing Fluency, can make a big difference in your customers’ behavior.

Processing Fluency

If someone speaks to you in a language you know very well (namely, one in which you are a fluent speaker), you will understand them better than if they spoke in a language you did not speak so well. This analogy holds for the idea of Processing Fluency: if you feel that you learned a piece of information more easily than another piece of information, then the first piece is more Fluent than the second. Because simple presentations are easier to understand, presenting the same information in a simple way versus a complex way will make it more Fluent.

This simple versus complex idea extends beyond the relative simplicity of a website. Text that is easier to read or easier to remember is usually more Fluent. For example, using italics is a good strategy for drawing attention to a specific word, but italicized words are also harder to read. Thus, writing a whole blog post in italics would make the entire piece less Fluent. Similarly, rhetorical devices such as rhyming makes information more memorable, and thus more Fluent.

15-bubblesSo what?

Why should you care about the Fluency of information? To start, it has several important consequences for customers’ choices and confidence, and can make the difference between whether they believe you or not. A great deal of research has looked systematically at these effects of Fluency (see here and here for reviews).

The most robust finding is that people tend to believe Fluent information more than they believe Disfluent information (e.g. information that is harder to understand). For example, Rolf Reber and Norbert Schwarz published a 1999 study that showed people are more likely to believe a false statement (such as “Lima is in Chile”) when it is easier to read (such as using dark blue text on a white background) than when it was harder to read (such as light blue text on a white background).

15-peruThe fact that consumers think Fluent information is more true is extremely important, and suggests that they are more likely to believe and acknowledge advertisements and product descriptions when they can understand them more easily. But while Fluency increases consumers’ certainty that information is true, it also increases their confidence that they have understood and learned the information. For example, a 2003 study found that people were more likely to think they could remember words when they were presented Fluently.

More interestingly, a 2007 paper by Adam Alter, Daniel Oppenheimer, Nicholas Epley, and Rebecca Eyre found that people were generally more confident that they had understood product information about an mp3 player when it was Fluent versus Disfluent, resulting in a few interesting observations:


  • First, participants exposed to a source of Fluent information often did not understand the information as well as those who saw the Disfluent information: they only thought that they understood it better. In other words, although simplicity helps understanding, it also can lead customers to overestimate how much they understand.
  • This lead to a second important observation: participants who saw the Fluent information were more susceptible to nudges in their environment (such as the Foot-in-the-Door or the Decoy Effects) because they were not thinking as carefully about the products. In other words, they were overconfident, and it hurt their ability to override their immediate decision-making intuitions.
  • Finally, when participants were more confident that they understood the product information, not only did they think it was more true, but they rated the products more highly.

This last observation suggests an even more important effect of Fluency, namely that Fluent information is also more attractive to consumers. A 2005 study by Petia Petrova and Robert Cialdini found that consumers overvalued products (specifically vacation packages) that they had an easier time imagining. In other words, when their beliefs about a vacation were more Fluent, they liked that vacation more. The authors also found that they could make consumers value the vacations more both by describing them in more vivid detail, and by simply telling consumers to imagine the experiences.

15-vacayMore starkly (and more surprisingly), a 2006 paper by Adam Alter and Daniel Oppenheimer found that stocks with more easily pronounceable ticker symbols tended to perform better in the short run because investors preferred them over less Fluently named stocks. In one of their studies, they looked at the stock performance of companies 1 day, 1 week, 6 months, and 12 months after they had been posted to the NYSE:

To emphasize just how successful investing in fluently named stocks would be, we calculated how much a $1,000 investment would yield when invested in a basket of the 10 most fluently named shares and the 10 most disfluently named shares. The fluent basket would have yielded a significantly greater profit at all four time periods: $112 after 1 day, $118 after 1 week, $277 after 6 months (all Ps < 0.05), and $333 after 1 year (P < 0.10).

In other words, new companies’ stock prices benefit from having a more Fluent ticker symbol, although the effect does go away in the long run.

Make it Fluent

Based on the research I’ve described here, making information more Fluent for your customers will make them believe it more, make them value it more, and make them use it more. So how do you make something more Fluent?

As I alluded to at the beginning of this post, one of the most important (and obvious) things you can do is to make sure customers’ environment (e.g. the website) has a simple design, is easy to use, and provides as little friction as possible at all stages of a purchase. On a more detailed level, make sure that the information to which you want customers to pay attention is printed cleanly and clearly, and you could even downplay the Fluency of information that you don’t want customers to see (such as by making it harder to read by putting it in italics or a faded color). Don’t forget that while less fluent information may just be downplayed in the presence of more Fluent information, sometimes it can also make customers feel like they don’t understand something, which reduces their confidence.

On a semantic level, you can also play with how easy it is to understand advertisements or product descriptions. Poetic and rhetorical devices make things more fluent, as does using more simple vocabulary. For example, many electronic products will have specs that the average consumer doesn’t understand, meaning that this information is Disfluent. Finding a way to rewrite these specs in a way that looks and feels more accessible will make customers more confident that they understand the information, and this can lead to greater interested in the product.

15-speakersFluency is very important because it affects how we form our first impressions of the information we learn, and first impressions are extremely important. However, there are actually many more ways Fluency can affect consumers’ decision making, for example, it also affects how familiar, famous, typical, and reliable products are perceived to be. Lastly, if someone’s writing is easier to process, readers tend to think the writer is smarter, so I hope you though this post was Fluent!


  1. The Fluency of information is important. It makes information seem more true, makes customers more confident, and makes them value products more.
  2. You can influence Fluency by making information easier to process, read, understand, or memorize.
  3. Bottom line: make things simple and memorable, especially the information you want customers to act on.

Guest post by Stanford’s Alex DePaoli

Follow him @AlexDePaoli.


The Blemishing Effect

It’s rare that a purchase has no downside. Even if the product is exactly what a customer is looking for, there may be some feature that doesn’t measure up as well to rival products. How bad is it if your customers are taking stock of the cons as well as the pros of your products? New research suggests that, as long as you manage how the negative information is encountered, it may actually be a good thing.

14-badnewsBlemishing leads to Blossoming

In a series of experiments, Danit Ein-Gar (of the University of Tel Aviv), Baba Shiv, and Zakary Tormala (both of Stanford Graduate School of Business) examined what happened when they presented consumers with a small bit of negative information about products they were already evaluating. This negative information ranged from actual negative product attributes (e.g. “only two” color options for a pair of hiking boots) to sources of inconvenience (e.g. a set of champagne glasses that didn’t come with a hard box) to very minuscule or irrelevant details (e.g. the box for a pair of shoes was scuffed).

14-boxThe authors found that under certain conditions (see below), having that little bit of negative information actually made consumers more likely to buy the product! In other words, telling customers that there was something mildly bad about the product made it more attractive. The authors call this the “Blemishing Effect.”

14-bootsWhy would customers like a pair of hiking boots more just because they were given a small piece of negative information about them? The authors suggest that it may be a byproduct of the way we evaluate information, and (more importantly) how we always like to be right.

Blemishing leads to Bolstering

The first and most essential thing to take into account about the Blemishing Effect is that it only works when the negative information is encountered after some stronger positive information. In other words, if you present the negative information first, or if the negative information is too negative, the effect goes away (and you may be worse off than if you had just never mentioned the negative at all).

14-scalePositive information needs to come first thanks to the Primacy Effect: arguments that customers encounter first will be more persuasive to them. This means customers can be made to develop a positive impression based on the initial positive information. Next, when they encounter a (minor) piece of negative information, it will lead them to very briefly reevaluate their impression of the product. Because the positives still outweigh the minor negative, their opinion won’t change, but the mere process of thinking about it and coming to the conclusion that the product is still good will make them like it more. When people arrive at a judgment on their own, they hold that belief much more strongly.

To illustrate how this process works, consider a 1979 paper by Charles Lord, Lee Ross, and Mark Lepper. The authors had 151 Stanford undergraduates read two studies about the effectiveness and merits of capital punishment. One of these studies “proved” that capital punishment was effective, while the other “proved” it was not. The authors had identified ahead of time which students were pro- or anti-capital punishment, and they used this information  to determine whether students would first read the study that agreed with their beliefs or first read the study that differed from their beliefs.

Their most important finding was that students did not treat both studies equally. Instead, they would use the study with which they already agreed in order to concoct arguments to undermine or discredit the study with which they did not agree (a demonstration of the classic Confirmation Bias). Furthermore, despite the fact that they had read one pro- and one anti-capital punishment study, they tended to emerge from the experiment with a stronger belief that they had been right the whole time. In other words, seeing arguments from both sides only made them double-down on their initial beliefs. The authors call this “Attitude Polarization” (and note that this is a heavily simplified version of their results).

14-confbiasWhat does this have to do with product evaluations? Essentially the Blemishing Effect works in the same way as Attitude Polarization: because customers need to think about the product, and decide that the minor negative information is not as important as the positive information, they will reinforce their positive beliefs.


Recall that I said the Blemishing Effect only arises under certain conditions. Here they are:

  • The positive information must outweigh the negative information.
  • The positive information must come before the negative information.
  • Consumers should not be too invested in the purchase decision.

As we’ve discussed, the Blemishing Effect arises because of natural biases in how customers evaluate information. However, this means that if they are willing to invest the time and cognitive resources to think carefully about the purchase, they can override these biases and the effect will go away (this is similar to many of the ideas in Daniel Kahneman’s book Thinking, Fast and Slow). If the negative information is very minor, this just means that you won’t get a boost from adding the negative (e.g. a scuffed shoebox won’t stop someone from buying shoes), but if it is more substantial, then it can actually hurt you (e.g. having a limited number of color options is undesirable).

14-RuffwearpngBlemishing leads to Buying

This suggests that the Blemishing Effect needs to be applied very carefully in the real world, and you may not expect to see the Blemishing Effect for products that people need to evaluate carefully. In other words, large, expensive, or otherwise important purchases probably won’t be helped by a Blemishing Effect. However, small or routine purchases might benefit from listing a small negative attribute (even if it seems irrelevant). Additionally, anytime you expect customers to be in a rush or otherwise distracted or depleted are times when you would also expect to see the Blemishing Effect arise.

Other than listing product attributes, there are two ways to blemish a product. One strategy is to call attention to a (mildly) unflattering product comparison. For example, having “two colors” as options for a pair of hiking boots is not necessarily seen as a negative until it is relabeled as “only two colors.” The second and perhaps simplest approach is by scuffing the packaging, or (in the case of an online retailer) you might experiment with the unintuitive strategy of using lower quality or less professional product photos.

14-boots2While Blemishing is a relatively delicate effect, some of takeaways from the research are still universal to all kinds of products. Notably, most products have features or attributes that are less appealing than others, so the research suggests that you should always list these attributes after you have listed some strong positive attributes. Putting the positive information first will still have the effect of making your customers’ first impression be something good.


  1. When listing pros and cons, always make sure you start with some strong pros before mentioning the cons. This will lead to customers advocating your product to themselves.
  2. If you have a product about which customers won’t be thinking too hard, include a small con (even if it’s irrelevant) in order to bolster customers’ positive evaluations of the product.

Guest post by Stanford’s Alex DePaoli

Follow him @AlexDePaoli.


Freshplum Covered In Forbes

Thank you Adam at Forbes Magazine, for your insightful look at how Big Data, coupled with Machine Learning, is changing the way retailers can better target their shoppers. One of the ways that Freshplum is providing this technology is by selectively targeting shoppers based on purchase intent. The results of this technology are clear, as evident from Ric Kostick, co-founder of retailer 100% Pure:

“’Kostick signed up. He knew some people flinch at paying $18 for an 8-ounce tube of shampoo. Perhaps occasional discounts would help. After three months using Freshplum’s selective promotions, online sales increased as much as 13.52%’, Kostick said.”

At Freshplum, we believe that Big Data can provide a richer experience to the customer by personalizing and targeting messages catered to their behavioral intent. By doing this, retailers can more easily convert shoppers when they’re browsing the site.


Have you ever felt that you have less self-control at certain times of day? It turns out the science of self-control and willpower has a lot to say about that. Understanding how consumers’ self-control works (and when it doesn’t) can actually allow us to predict what time of day people are more likely to indulge the urge to splurge.

Ego Depletion

The most popular scientific model of self-control and willpower is that of a limited resource. In essence, whenever you choose to exercise self-control (for example, by choosing to skip dessert or to read a book instead of watching a TV show), it uses up some of your “cognitive resources.” This means that the more you exercise your self-control, the harder it will be to exert your willpower later on. If you choose to eat healthy instead of unhealthy all day, and to exercise instead of sitting around, you’re more likely to lose the self-control battle to a piece of cake later that night. And because willpower is a single resource, this also means you’ll have a harder time resisting a tempting impulse purchase.

13-saladcakeThis idea of self-control was originally championed by Roy Baumeister, and he refers to this process of draining your cognitive resources as “Ego Depletion.” As consumers make choices, they experience Ego Depletion and subsequently have a harder time exercising their self-control. However, cognitive resources can be replenished: taking a rest, eating a meal, having fun, or being mindful of your depletion can all help restore your cognitive resources.

13-muscleIn order to better describe this idea of a limited but refillable resource, Baumeister often compares willpower to a muscle. The more a consumer flexes their willpower muscles, the more fatigued the muscle gets. However, this fatigue is only temporary. Also not unlike a muscle, it is possible to strengthen one’s willpower with practice.

Weakening Willpower

When customers are depleted, they are more likely to indulge and more likely to make a purchase. They are also more susceptible to nudges in their environment, so they will experience things like the Foot-in-the-Door Effect, the Decoy Effect, or the Primacy Effect. As marketers, this leads us to ask: when and how do customers become depleted?

Most obviously, any time a customer must use their willpower to make a choice that sacrifices short-term enjoyment for the sake of a longer-term goal, we can say that they have exercised self-control and will experience some Ego Depletion. However, there are actually many things that can drain a person’s cognitive resources, most prominently feeling physically exhausted, experiencing negative emotions, or doing something that is cognitively taxing.

13-dilbertThis last point is especially noteworthy. Essentially, when people need to think long and/or hard, it will use some of their cognitive resources. This means that if consumers see your ads while they are doing something that requires a lot of mental work, they will probably be more open to your message. In a retail setting, this means that if customers are asked a series of questions that make them think, they will be more Ego Depleted by the time you ask them if they want to buy a product (which is very similar to a 2009 study run by Bob Fennis, Loes Janssen, and Kathleen Vohs). There are a couple of ways this could be done in an online context.

One opportunity is to present your visitors with gamified choices requiring them to make difficult product trade-offs by presenting them with a pair of products and asking them which they like more. Not only could this build engagement while leading to Ego Depletion, but it may allow you to run some real-time market research. Alternatively, similar to the Fennis, Jansen, and Vohs (2009) study, you could ask them relevant survey questions that they can answer if they wish. Using metadata, you can tailor these survey questions to increase the level of interest for individual customers, and thus increase the likelihood of them spending the cognitive resources necessary to answer.


13-clockOne especially interesting result of Ego Depletion is that it allows us to predict that consumers will (in general) have less self-control later in the day. In other words, rest is one of the factors that effectively replenishes willpower, so people tend to have good self-control in the mornings. However, as the day goes on and they make decisions or engage in work or thinking, they experience Ego Depletion, leaving them with fewer cognitive resources later in the day. The obvious conclusion is that your potential customers will be more interested in your products (or at least more likely to buy them) later in the day.

13-snacksThere is an important caveat to this assumption, namely that eating a meal can also serve to replenish willpower. A 2008 study by Roy Baumeister and E.J. Masicampo found that blood glucose levels tended to be positively associated with willpower, meaning that in a very direct way, eating a meal serves to restore cognitive resources.

In an even more stark demonstration, a recent article by Shai Danziger, Jonathan Levav, and Liora Avnaim-Pesso looked at Israeli court decisions in which a single judge would listen to several appeal cases back-to-back. They found that judges who recently had something to eat were far more likely to listen to a defendant plead their case, and that the judges’ patience (and willingness to consider the individual facts) dropped off dramatically between meal or snack breaks.

The graph below shows these findings. Note how the proportion of favorable rulings are highest at three points: after breakfast, after lunch, and after an afternoon snack (indicated by the circles). Also note that, as we predicted earlier, judges’ willingness to consider the cases goes down dramatically from the morning to the afternoon.

13-levavSo when are consumers most likely to have low willpower? The research suggests that it will be later in the day, before a meal, and when they have been working or thinking hard. While each piece can be used to target different parts of the day, taken together we would predict that the end of the workday is when people are most susceptible to the urge to splurge.


  1. Self-control is like a muscle: if customers use it too much, it gets exhausted and they are more likely to indulge.
  2. Consumers’ self-control can get exhausted when they make a lot of choices and when they think long and/or hard about something, effectively using up their limited cognitive reosurces.
  3. Customer self-control will be strongest early in the day and right after meals.

Guest post by Stanford’s Alex DePaoli

Follow him @AlexDePaoli.

{ 1 comment }

Shopping Momentum

Here at the Freshplum Blog, we’re big advocates of giving gifts to your customers. Whether to promote a sense of reciprocity or to create a feeling of endowment, getting something in your customers’ shopping carts is a great way to encourage them to shop with you. However, it turns out there’s an even better way to get the ball rolling than by giving gifts: make your customers pay instead.

Shopping Momentum

Normally we think of getting customers to buy something as the end goal rather than the starting point, but just putting your customers into the mind-set of buying rather than browsing is incredibly powerful. Illustrating this, a 2007 article by Ravi Dhar, Joel Huber, and Uzma Khan (of the Yale, Duke, and Stanford business schools respectively) documented an effect they dubbed “Shopping Momentum.”

12-momomentumThe Shopping Momentum Effect describes something that most impulse buyers may already suspect to be true: consumers who choose to make a purchase are subsequently much more likely to make another, unrelated purchase. In one of the authors’ examples, participants first chose whether or not to buy a light bulb, and then chose whether or not to buy a keychain. If they did not buy the light bulb, there was a 31% chance that they would choose to buy the keychain, but if they did buy the light bulb, then the likelihood of them buying the keychain more than doubled to 67%!

In another example, the authors compared the effect of Shopping Momentum to the effect of giving customers a gift. Participants were either given a pen as a gift or asked if they wanted to buy a pen, and then they were given the chance to buy a keychain. If they received the pen as a gift, 53% chose to buy the keychain, which is not bad at all. However, if they had the opportunity to buy the pen, 78% chose to buy the keychain!

Implementation Mind-Sets

Why are customers more likely to buy something after buying something else? Conceptually, this is similar to a Foot-in-the-Door Effect, namely that once your customers overcome the hurdle of buying their first product (or opening an account or at least placing something in their cart), they are much more comfortable buying another one.  However, according to the Dhar, Huber, and Khan paper, the story is a little bit more interesting.

Building on work by Peter Gollwitzer, they propose that there are fundamentally two ways customers think about potential purchases:

  • They can be in a deliberative mind-set, in which they evaluate the product and think about its pros and cons
  • They can be in an implementation mind-set, in which they think about what they need to do in order to actually make a purchase

If your customers are thinking about your products and your deals in a full implementation mind-set, then they are focusing on the “when” and “how” of the purchase instead of the “why”, and this makes them much more likely to buy something.

12-thinkdoHow does this relate to Shopping Momentum? The authors show that when customers buy the first item, it pushes them out of a deliberative mind-set and into an implementation mind-set, making it more likely they will buy other products too. More interestingly, they showed that by simply putting consumers into an implementation mind-set using an entirely different method, they could produce a Shopping Momentum effect without the participants even making an initial purchase. For example, if participants were asked to write down the steps they would need to take in order to buy a new car (which got them thinking about implementation), then 66% of them opted to buy a keychain.

Building Momentum

The Shopping Momentum Effect can be applied to retail strategy at two different levels.

The first strategy is to apply Shopping Momentum as simply the effect of buying something. Deals, promotions, or strongly tempting loss-leaders can serve as ways to get customers to buy something, which in turn will make them more likely to buy something else. Offering little things for a dollar that you could have otherwise given to them as a gift might also work, although the uptake will probably not be high.

In addition, anything you can do to reduce the friction customers experience during the buying process will be very helpful. They are much more likely to follow through with that initial bait purchase if the process of buying and paying is very easy (such as with Amazon’s One-Click functionality). And of course, once they have made a purchase it needs to be easy for them to see and buy other products.

12-1clickThe second strategy is to play with the underlying cause of Shopping Momentum, namely the implementation mind-set. One way might be to highlight the steps needed to purchase a product in the space where customers might normally expect to see product information (with that information moved further down the page). By putting this implementation information front-and-center, your customers will be more likely to start thinking in these terms.

You could do something similar to the study, namely ask your customers to consider the process they need to follow in order to buy something (they don’t necessarily need to give you an answer), or more subtly, ask them more innocuous implementation questions such as “When do you need this by?” or “How do you plan to pay?”


This ties in nicely with research demonstrating that one of the most potent ways to encourage people to follow through with a behavior is to encourage them to think about a plan of action. Beginning with a 1965 paper by Howard Leventhal and colleagues, this same idea is now frequently applied by political campaigns to encourage constituents to go to the polls. Likewise, any wording that encourages your customers to think about how they can buy your product may make them more inclined to do so.


  1. Your customers are much more likely to buy something if they have already bought something else.
  2. This is because they are more likely to buy something if they are thinking about following-through with a purchase rather than thinking about the pros and cons of a purchase.
  3. Encourage your customers to start thinking about how to buy your products.

Guest post by Stanford’s Alex DePaoli

Follow him @AlexDePaoli.




Crafting a compelling message can be one of the biggest challenges in marketing. How can you motivate your customers with only a few words in an ad or just a few phrases in a product description? While there are countless approaches, one universally useful tip comes out of the study of the psychology of goal pursuit: to make a strong appeal, you need to match how consumers expect to use a product to the “direction” of their motivations for using it.

11-apav“Get Excited” or “Don’t Miss Out”

Suppose you are marketing athletic apparel, and you want to appeal to customers’ desire to go to the gym. The average person wants to go the gym for weight-loss and/or fitness reasons, but they may think about their goal(s) in different ways. On one hand, a customer may want to get fit, while on the other hand they may want to lose weight. These may be effectively the same thing, except one of them is presented as gaining fitness, while the other is oriented towards avoiding fatness.

This distinction is based on the long-running research of Columbia University’s E. Tory Higgins (this paper from 2000 offers a nice overview), who has suggested that people tend to be motivated either through a Promotion-focus (e.g. “how can I achieve a good outcome?”) or in a Prevention-focus (e.g. “how can I avoid a bad outcome?”). The endpoint of these two approaches is the same (for example, “to go to the gym”), but the motivations are opposite (to “be fit” or to “not be fat”). Basically, if customers are hoping to reach a positive outcome, then they are in a Promotion-focus (“I want to win!”). If they are concerned about getting away from a bad outcome, they are in a Prevention-focus (“I don’t want to lose!”).

11-shirtHiggins refers to the adoption of a Promotion- or Prevention-focus as a person’s Regulatory Focus, and when a product or advertisement fits with a customer’s Regulatory Focus, they are more interested in it and more motivated by it. For example, if people want to “get fit,” then they are more motivated by ads that are phrased in terms of potential achievement, while people who want to “not be fat” will be more motivated by ads that emphasize worst-case scenarios.

When there is a mismatch, consumers tend to be more skeptical or otherwise less interested. For example, if someone thinks about going to the gym as a way to “get fit” (or is accustomed to athletic wear ads that are inspirational or otherwise Promotion-focused, such as Nike’s “Just Do It” campaign), they may be confused or even offended to encounter a Prevention-focused ad that stresses “don’t be fat.”


Promote or Prevent?

What determines which Regulatory Focus a customer will have? In general, some people are more inclined to think about their goals with a Promotion-focus while others naturally tend to adopt a Prevention-focus. However, more often their Focus is context specific, either because people tend to think about certain products in a certain way, or because you can actually influence how they feel.

11-welchsAngela Lee and Jennifer Aaker (co-author of The Dragonfly Effectdemonstrated that if you know whether your customers tend to view your product as Promotion- or Prevention-focused, then you can increase engagement by tailoring your message to fit their perspective. In one of their studies, for example, they had participants read about the benefits of drinking grape juice. Half of the people read about how juice provided a boost of energy that helped you do whatever you wanted (a Promotion-focused attitude towards the product), while the other half read about how juice helped prevent cardiovascular disease to keep you healthy (a Prevention-focused attitude). When asked to evaluate an ad for the juice that was phrased either as “Get Energized!” (Promotion) or “Don’t Miss Out on Getting Energized!” (Prevention), people liked the ad a lot more when it matched their current focus. In other words, if they were led to believe that the juice was meant to energize them, they wanted to “Get Energized!”, but if they thought the juice was meant to keep them safe and healthy, then they didn’t want to “Miss Out on Getting Energized!”

Just about any product goal can be presented as achieving something positive or avoiding something negative, and both can be strong motivators. For example, products as diverse as acne medication to clothing can be billed as “Look beautiful” or “Don’t miss out on being beautiful!” Product or app upgrades can be described as “improving performance” versus “fixing bugs.” Cell phone carriers can claim to have “More coverage!” versus “Fewer dropped calls!” Even a microwave could be sold as “Hot food in seconds!” versus “Never eat cold leftovers again!”

11-clothingadDetermining how how your customers tend to think about your product (or how you want to present your product to them), and then matching their expectations in your product’s presentation, can help you increase interest and engagement.

You’ve got to Focus!

The first step to applying Regulatory Focus is to simply ask (either yourself or your customers): what does this product do? More specifically, you want to know what is the most natural (or most common) way to describe the effect of the product.

Does it make the most sense to think about the product as helping customers get something positive or good? Do they want it in order to achieve or to fulfill a goal? If this is the case, then using a Promotion-focused message (that emphasizes approaching or reaching positive outcomes) will resonate better with your customers. For example, people tend to buy junk food because they want to gain the enjoyment of eating it, not to avoid some consequence of not eating it. This may be why Coca-Cola sticks to slogans like “Open happiness” or “Life begins here,” while not mentioning anything Coke might help avoid or fix (such as “thirst”) since 1939 (“Coca-Cola has the taste thirst goes for”).

11-cokeAlternatively, does it make sense to think about the product in terms of helping customers avoid negative outcomes or pain points? Do they want it in order to avoid or resolve a problem? If this is the case, then using a Prevention-focused message (that emphasizes avoiding or counteracting negative outcomes) will resonate better with your customers. For example, people tend to buy medications while sick or injured in order to get rid of an unpleasant feeling, so it is no surprise that Tylenol has the slogan, “Feel better.”

11-tylenolIdentifying how your customers think about your products (or perhaps how you want them to think about it) can inform how you promote and advertise. Aside from slogans, product features can be rewritten in order to reflect or appeal to the target Regulatory Focus, and advertisements can be worded to reflect this as well.

Recall that Regulatory Focus can be context specific. A more risky idea might be to actually use your Promotion or Prevention descriptions to create your own psychological manipulation. According to a 1998 paper by James Shah, E. Tory Higgins, and Ronald Friedman, Regulatory Focus interacts with Loss Aversion: if you have  a Promotion-focused message, suggest that the customer is getting a great deal because they are gaining the quality or the value of the product, whereas if you are using a Prevention-focus, suggest that the customer is getting a great deal because they are saving a lot of money or reducing a lot of risk.


  1. People are either motivated to achieve positive outcomes or to avoid negative outcomes. Match your message to your customers’ motivation to increase interest.
  2. If customers are in a mindset to achieve positives or avoid negatives, this focus carries over to how they evaluate your ads, your products, and your deals.

Guest post by Stanford’s Alex DePaoli

Follow him @AlexDePaoli.


The Primacy Effect

One of the simplest questions to ask about consumer behavior is “does order matter?” In other words, does the order in which you present things to your customers affect their decisions? It may or may not surprise you, but yes, order matters. In fact, order matters for a lot of things.


First of all, memory is sensitive to order. People are more likely to recall words (or any information) from a list when they encounter them toward the beginning or toward the end of the list, a phenomenon first described in 1913 by Hermann Ebbinghaus.

There are two separate effects of order: “Primacy,” in which people pay more attention to the first thing they read or hear, and “Recency,” in which people are (surprise!) better able to remember more recent information.

While Primacy and Recency both occur for memory, this is not always the case for decision making.

When people are choosing from a list, they often display a strong Primacy effect. In other words, if something is earlier in the list, people are more likely to choose it. It doesn’t matter if the list is ordered top-to-bottom or left-to-right, and it also (usually) doesn’t matter if people see the whole list at once or one item at a time. If you want people to click on something, put it earlier in the list.

An important exception to this rule of Primacy is that it only arises when people are reading their options. If people are listening to their choices, then they are actually more likely to pick the last option. Obviously most online decision making is done through reading and clicking, but if you ever decide to incorporate more sound or video into your website, don’t forget to rely on Recency!

It’s Everywhere

The fact that we are more likely to choose the first things they read can impact a lot of what we do. One of the more concerning consequences of Primacy was documented in a 1998 paper by Joanne Miller and Jon Krosnick (then at Ohio State). The authors looked at results from the 1992 elections in Ohio, and examined 118 individual races. In half of those races, particularly in counties where interest in politics was lower, the candidate listed first tended to get 2.5% more of the total vote.

In the context of online advertising and SEM, naturally it is always desirable to be high in search results to increase visibility. However, it also helps to appear higher on the list of ads that Google shows on the right-hand side of the search results. As people work their way down the list, they will be more interested in the first few items.

10-googleThese results give a good hint for when and why people are more likely to choose the first option. One of the most important (and obvious) things to keep in mind is that customers with a strong preference will select their preferred option no matter where it is in the list. Primacy arises for people who aren’t sure exactly what they want, and in this case one of two things can happen:

On one hand, people may not be interested in making the decision and will simply choose the first thing presented to them because it is easiest (unfortunately, this is probably the explanation for what happens during elections).

On the other hand, people may examine their options in order to make a good choice, but will naturally get cognitively tired as they go. Because they start with the first option, they end up thinking about the first option more than any other option. This leads them to feel like they understand the first one best, and it becomes the most comfortable choice (note that most people often don’t realize when they are doing this). A good example of this comes from a 1969 study by Vernon Stone in the psychology of law: mock jurors who read prosecution arguments before defense arguments tended to vote “guilty,” and vice versa.

coffeemugThis second explanation is important because it suggests that the order in which people consider information influences their decisions. In other words, people naturally place more weight on the first pieces of information that they see and think about. In a 2007 paper, Eric Johnson, Gerald Haubl, and Anat Keinan found that people who thought first about things they liked about a product (in their case, a coffee mug), instead of thinking about the price of the product, valued the product more. Because they had considered the positive qualities first, they unconsciously felt that these positive traits were more important.

Control the Order

The tendency for customers to show a Primacy effect lends itself to a number of possible applications. The most obvious use is to put an item at the beginning of a choice set or list when you want people to choose that item. This could be for pushing certain products, in which case it would be a great way to strengthen other choice effects (such as improving the Compromise Effect by placing the compromise option first), for getting customers to endorse a statement to which you want them to stick (as I discussed in my post on the Foot-in-the-Door Effect), or for getting higher click-through on a specific online advertisement.

10-describeuselfMore subtly, Primacy also has important implications for any market research you might try to conduct with your customers or clients. If you have a survey or are asking multiple-choice questions, the Primacy effect may lead to you getting answers biased towards the first choice. To fix this, you should randomize the order of choices in any situation where you want unbiased results.

Equally as important as controlling the order of options is controlling the order of information. It is not uncommon to see an advertisement present its strongest argument in the biggest letters for this exact reason: if people notice it first, they will process it first and give it more weight.

Why can’t this be done with an online retail layout? The main points of value for each product can be made especially visible, while price is relegated to a smaller font or may not even be visible without clicking through to a product page. Note that this is the exact opposite of what many sites do, which is to display price upfront but require customers to click through in order to see the details that will make them want to buy.

10-mugsIf your customers see the positive details about your product before they have to evaluate the price, they will be inclined to value it more highly.


  1. If you want more customers to choose a specific item (out of a list, choice set, or anything else they are reading), put that item first in line.
  2. Make sure your customers see positive information before negative information because they will put more weight on whatever they process first.

Guest post by Stanford’s Alex DePaoli

Follow him @AlexDePaoli.


The Foot In The Door Effect

One of the most well known findings in the psychology of marketing and persuasion is the “foot-in-the-door” effect. In its original and most famous demonstration, Jonathan Freedman and Scott Fraser went to homeowners in Palo Alto and asked them if they would be willing to put up a large, ugly sign in their front yards that said “Drive Carefully.” Only 17% were willing to put up such a sign.

However, for a second group of houses, Freedman and Fraser tried something more subtle. The homeowners were simply asked to place a small, 3 inch sign in their windows advocating safe driving, a request designed to be so innocuous that nearly everyone would agree to it. Two weeks later, a different person showed up at these same homes and asked if they would they be willing to put up the large, ugly sign. This time, 76% of homeowners agreed to put it up!

 9-impactAs the name “foot-in-the-door” implies, this effect occurs when customers have done something small for you, which leads them to be more willing to do something bigger for you later on. It is also a well known and widely used tactic in everything from auto sales to telemarketing (where it dramatically increases people’s likelihood to stay on the phone). However, using it in an online context has some unique challenges, so in this post I will outline some ways to effectively get an online foot-in-the-door.

Commitment and Consistency

There are a lot of factors that can drive the foot-in-the-door phenomenon depending on context, but at its core is an idea that Robert Cialdini calls “commitment and consistency.” Essentially, a homeowner or customer’s first action serves as a point of commitment to a behavior or attitude. When it is time to make the second decision (e.g. to place a large sign on your lawn or to purchase something online), the fact that your customer has already committed (even in a costless way) to the first action leads them to feel a need to be consistent with that commitment.

An important consequence of this mechanism is that the first action must be voluntary: if a customer is forced to do something, it will not serve as a foot-in-the-door because they do not feel a need to be consistent with it.

Naturally a number of other factors will affect the strength of foot-in-the-door as well. For example, if the first action is more public or visible, if it was more personally relevant, or if was expensive (in money, time, etc), customers will be more likely to commit to it.

If I Did it I’ll Do it Again

Because past behavior is such a strong first step for a foot-in-the-door effect, anything that draws customers’ attention to relevant past actions has the potential to open the door. One of the most obvious implications is that reminding returning customers of their previous purchases can be very powerful. This can come in a couple of forms.

One example would be a sidebar that shows either a list or a tally of previous purchases. This feature could be designed to “help customers keep track of past purchases” or to encourage them to rate products, but it would also serve to reinforce their relationship with your website as a place where they go to buy things. “I’ve bought these things from here before, so I must trust the site enough to buy again.” (Potentially, such a sidebar could even be designed to gamify how customers add new purchases to the list.)

A second approach would be to strengthen the message accompanying a recommendation engine. To use examples from Amazon, recommendations ought to use fewer messages like “Recommended for you” and more messages like “Customers who bought this item also bought.” General recommendation pages may also evoke specific past purchases: “If you liked X, then you’ll also like Y.”

Another approach might be to entice a customer to add something to their cart in order to jump-start their willingness to shop. For example, when a customer arrives on the site, they may be greeted with a message and a choice: “Welcome to our website! Please choose a free gift…” This approach not only encourages customers to engage with a product and with their cart, but it also leverages reciprocity.

9-freegiftAsk Me Anything

It is also important to note that the foot-in-the-door effect is not limited to behaviors that are similar to the behavior you are hoping to create. In the Freedman and Fraser study, while they found that asking for similar behaviors (i.e. agreeing to a small sign and to a big sign) produced a 76% compliance rate, they also found that asking for an initial behavior that was not the same (i.e. signing a petition or agreeing to a small sign that advocated a different cause) still led to a 48% compliance rate.

This suggests that a foot-in-the-door effect can be created by prompting visitors for input other than purchases. For example, simple and costless calls to action can lead your customer to start engaging with your site, which may serve as a foot-in-the-door for engaging with the products. Just ask them a question or prompt them to click on something, just try to make it risk free, fun, and maybe a little meaningful. For example, asking your customers about their preferences (like Netflix does for new accounts) might serve as a valuable way to get a recommendation engine started (or to design messages for customers with no purchase histories: “If you like action movies, then you will like…”), but getting them to tell you what they like (or want) may also serve to get a foot-in-the-door.

Conversions Aren’t Everything

Up to this point, the focus of this post has been on encouraging purchasing behavior by first getting customers to do something with your site, but the foot-in-the-door effect is perhaps even better suited to encouraging relatively more costless behaviors. For example, if you have a mailing list that you want visitors to join or a source of content marketing you want them to view, prompt them with a simple pop-up or banner question. If they respond (or when they respond depending on how you set it up), prompt them to sign-up or follow a link with the understanding that doing so would be consistent with the opinion they just expressed. The content-sharing site Upworthy had implemented pop-ups that did exactly this:


  1. Get customers to voluntarily do something small for you in order to make them more likely to do something big for you later.
  2. Get them to express their opinions, values, or preferences (either by asking or by observing), and then remind them of these as they shop.

Guest post by Stanford’s Alex DePaoli

Follow him @AlexDePaoli.

{ 1 comment }

With relatively few exceptions, sales events are very important in retail. The reason is obvious: people love to get good deals. Or at least they love to feel like they are getting good deals.

8-saleThis post will explore some of the behavioral economic underpinnings of how people respond to good deals. As the experience of buying something on sale is common to nearly everyone, I expect you will find several observations to be very intuitive, but I hope to get you thinking about them in some novel ways.

Reference Prices

Here’s a sleazy trick you may have seen before. A store is having a huge sales event, and everywhere you look are big signs where the original price is slashed and the new price is tantalizingly low. Except it isn’t really, because that “original price” is artificially high. Likewise you may have noticed that the “suggested retail price” of some products is often undercut by retailers, suggesting that it too may be higher than it needs to be. Both of these tricks are meant to make you feel like you are getting a good deal, which any retailer or salesperson knows is extremely important to most customers.

There are a broad range of ways to influence how much a customer will like a deal. For instance, a stream of research by Margaret Campbell found that people are more likely to think they getting a great deal when they are in a good mood or when they have a high opinion of the store from which they are buying.

In a 1985 paper (and in a more readable 1999 paper), Richard Thaler (co-author of Nudge) suggested that people tend to be almost as interested in “buying” a deal as they are in buying the associated product. In other words, when customers are deciding whether or not a product is worth the price (for example, an expensive pair of shoes), the value they place on the purchase itself (“Am I getting a good deal?”) can be as powerful as the value they place on the actual product (“Am I getting good shoes?”).

The most direct way to influence perceptions of a transaction is through careful use of reference prices. Basically, it is important to present your customers not only with the price they will be paying, but also with a relevant price that makes the real one look good by comparison. This is the purpose of posting the original price during a sale. The obvious takeaway is that whenever your product is discounted, make sure your customers can see what the price was before you slashed it. Don’t just tell them “It’s X% off!”

8-zappossaleOf course there are any number of tricks that can be played with reference prices, and they don’t need to be sleazy. For example, if your firm competes on price, then drawing attention to the higher prices of competitors’ products is a well-established and powerful tactic. Taking this a step further and treating the competition’s price as a reference, if your potential customers have the sense that your competitor is charging a “normal” or “usual” price, your cheaper product will be seen as an even better deal. Even if your prices do not differ from your competitors’, offering even a modest discount and then comparing the new price to the competition will be especially persuasive.

Another fruitful approach is to use the prices of similar or related products to make the price of your target product look good. For example, even if they are not on sale, a $125 pair of shoes look like a great deal next to a $190 pair of shoes.

 8-zapposThis type of reference has an added bonus: presenting your customers with similar products will also lead them to pay more attention to the deals you offer them. A brand new paper by Christina Kan and colleagues explored this approach, and found that shoppers were more interested in how much they can save when a product was advertised alongside similar products.

Using reference prices

Suppose Zappos wants to encourage sales of a given brand of dress shoe. They could use either strategy shown above: they can offer a sale and display both the old and the new prices, or they can display and advertise the shoe as part of a favorable comparison. They could also do both:

8-zappossale2Or for bonus points, they could combine this idea with suggestions from the Compromise or Decoy Effects. Below is an example of using a $99 pair of shoes as a decoy:

8-zappossale3Remember that the use of reference prices is not limited t0 products that are actually on sale. The essential thing to remember is that customers are very sensitive to comparisons, so the more that you can present them with favorable comparisons, the more they will think of your store as a place they can get good deals.


  1. People make purchasing decisions based on how much they value the deal, not just how much they value the product.
  2. Whenever possible, give your customers a reference price that makes your deal look good by comparison.
  3. Make sure the reference price is relevant, whether an old price, a competitor’s price, or the price of a similar or related product.

Guest post by Stanford’s Alex DePaoli

Follow him @AlexDePaoli.