Why It Works—Behavioral Economics, Psychology, and Game Theory
More than Meets the Eye
FairPay draws on elements of pay what you want (PWYW) pricing—that simple strategy has gotten some attention in the past few years, but most people only dimly understand it. A growing body of research and actual usage is showing that it has much more potential value than most businesses realize. People actually do pay even when they do not have to—often generously.
For those who think there can be no real money in PWYW, the success of Humble Bundle shows that to be a misconception. They started with limited time offers of bundles of indie games, and later expanded to e-books. Their model includes a combination of payouts to developers and to charities. As of December 2014, it was reported that they had generated over $100M dollars for developers of games and other content sold in their bundles (since 2010), plus more than another $50M in additional customer revenues passed to charity. (In 2011, they raised $4.48M in funding from major venture capital firms.) One of the charities they support, the Electronic Frontier Foundation posted a nice summary (Kamdar 2011), observing that:
While the record labels, movie studios, and video game producers have not figured out a way to compete with free, others have ... as the Humble Bundle has shown us, it is possible, with creators and distributors finding new and exciting ways to compete with free. ... when done right—developers, content providers, and even those who provide the business model can successfully compete with free.
There are many other real-world examples of PWYW successes in promotional sales—for both digital and real offerings—as summarized in the following sections (details and references at FPZLink).
Research on Making Customers Want to Pay You
Classical economics suggests that customers will pay zero if given a conventional PWYW offer (since payment is entirely voluntary). But recent business experiences and behavioral economics research suggest that thinking is just stuck in the mindset of the last century.
Modern mass-market commerce is a race to the bottom that assumes and appeals to the worst in people. Sellers set prices as high as they think they can to maximize total profit—so buyers’ only option is to take it, or bargain-hunt. What we have here is what behavioral economists call an exchange relationship norm. Exchange norms are zero-sum, quid-pro-quo. Conventional PWYW changes that relationship in important ways—and FairPay does so far more deeply.
Insight into how and why real consumer pricing behavior is more generous than classical economics would have us think, emerges from a growing body of research and real business success with PWYW pricing. I summarize the key lessons in terms of a two-dimensional view of behavior that is my interpretation of a paper by Santana and Morwitz (2013), aptly titled “We’re in This Together: How Sellers, Social Values, and Relationship Norms Influence Consumer Payments in Pay-What-You-Want Contexts”:
Social Value Orientation (SVO), essentially pro-social versus pro-self—as individual traits.
Economic/Exchange Relationship Norms versus Communal Relationship Norms—as situational variables in a relationship.
This provides the motivation for the two-dimensional strategy that FairPay seeks to apply to getting buyers to willingly pay a fair price:
Segment customers based on their Social Value Orientation traits—are they receptive to and driven by social values, or not? Tactics for managing the FairPay process will be a bit different for high, medium, and low SVO segments.
Nudge all customers toward Communal Relationship Norms, in ways tuned to each segment—to seek to bring out their Social Value Orientation to the fullest extent possible.
Based on this, the FairPay process can be understood to work for each segment, but with rather different control parameters applied to each. In all cases the objective is to foster a situation that favors Communal Relationship Norms, and that draws out whatever level of Social Value Orientation can be elicited.
The sweet spot is targeting high Social Value Orientation (pro-social) customers, and moving them toward Communal Relationship Norms. They are the ones who will respond best to the pricing privilege that the seller grants to the buyer in FairPay—to price in a way that considers fairness to the seller—and who will be least inclined to abuse that privilege. Managing that for these buyers will be mostly carrot, and not much stick.
A secondary focus is on moving medium-to-low Social Value Orientation (more pro-self) customers toward Communal Relationship Norms. They will need more nudging to emphasize the carrot (why the seller is deserving of communal norms), while keeping the stick in sight (why it is in their best interest to price fairly). Those who do not respond with at least minimum levels of fairness (uncooperatively pro-self) can be treated as a third segment—to be excluded from FairPay (at least until they seem ready to behave more sociably), and be left to buy on the conventional set-price terms that routinely work for pure Exchange Relationships.
Businesses can seek to maximize profits with a mix of FairPay and conventional set-pricing by doing the following (as explained throughout this book):
Position themselves as deserving of Communal Relationship Norms. This can cover the whole spectrum of corporate citizenship, customer relations, quality, style, artistry, crafts-manship, service, and support.
Sustain that positioning throughout their customer relationships. This is deeply embedded in the FairPay processes.
Seek to market to high SVO (pro-social) customers as the preferred market segment. This is the segment that will be most willing to pay you generously for your product or service (if you position yourself as deserving, and ask in the right way).
Manage the segmentation throughout the business processes to appeal in the right way to the right people. FairPay provides an architecture that supports this. In contrast, freemium has been very popular because of its simple segmentation between those who pay and those who don’t, but has been limited in its ability to optimize and up-sell that, as explained earlier.
Such strategies for finding the sweet spots are especially important to targeting early uses of FairPay (as emphasized in Chapter 14). Given the learning curves and the risk that ill-targeted trials might lead to unsatisfying results, it is important that initial uses focus on customers who have high (pro-social) Social Value Orientations, in contexts in which they are appreciative and loyal, and so can be motivated to lean toward communal norms. That will favor the generosity and cooperation needed to make FairPay work well. Good results in such contexts will create a base for building out more broadly from there.
A further insight from this study is to reinforce that the nudging of buyers in the adaptive control process of FairPay is best done with a gentle hand. Communal Relationship Norms are a delicate thing. There is in FairPay an inherent quid-pro-quo over time (in the future, those who do not pay well will get fewer and less generous offers than those who do pay well). This should be managed with enough flexibility, generosity, and forgiveness of minor lapses, to reinforce, not poison, the effort to nudge toward Communal Relationship Norms. Much useful background is in the paper (as highlighted at FPZLink).
Making Pay What You Want Profitable and Sustainable
I summarized some specific points applicable to both simple PWYW, and the more sophisticated approach of FairPay when interviewed by Tom Morkes, author of “The Complete Guide to Pay What You Want Pricing” (2013). That rich and useful guide is full of insights on best practices for making PWYW work that also apply to FairPay. PWYW draws on subtleties in human behavior—it can be very powerful, but there is nuance to framing such offers, and making it profitable on a sustained basis has been a challenge.
Tom provides a nice PWYW checklist. He lists 11 steps to making a PWYW offer work. I group Tom’s list as follows, and add one more:
Numbers 1 to 4 are prerequisites to using these methods, both PWYW and FairPay:
Identify a competitive marketplace
Identify and target a demographic with fair-minded customers
Determine a product with low marginal cost
Create a product that can be sold credibly at a wide range of prices.
Numbers 5 to 11 enhance the process of using these methods, also relevant to PWYW and FairPay:
5. Establish a strong relationship with your customer
6. Clarify the offer
7. Show the customer that you’re human (even large corporations can emphasize their human values and the people behind them)
8. Appeal to idealism
9. Anchor the price
10. Steer the customer to the right choice
11. Remind your audience to contribute.
To which FairPay adds a new 12th step, to roll it into a cyclic, adaptive, individualized process that is ongoing (a repeated game):
12. Repeat offers contingent on fairness—build continuing relationship and dialog.
Online Resource Guide to Pricing
I have posted a “Resource Guide to Pricing—Finding Fair Value Exchange” (Reisman 2016) as a partial survey of emerging research on PWYW and related pricing strategies. Some key points:
A whole new era in pricing models is just beginning, spurred by the capabilities of the Internet. A nice summary of some of these directions is in the book Smart Pricing by Raju and Zhang of the Wharton School (Raju and Zhang 2010), which has illuminating chapters on PWYW and many other innovative models. Two other notable books on pricing are also listed (Anderson 2009; Nagle, Hogan, and Zale 2010).
The Egbert, Greiff, and Xhangolli (2014) paper is particularly relevant to FairPay, as one of the few that highlights the value of post-pricing. This key element of FairPay has been under-recognized (when not missed completely) in most research and trials of PWYW.
Others specific to PWYW are a series of research papers published since 2009—all the 12 included at this writing find it quite effective in a range of situations, revealing a number of reasons why people actually chose to pay, and to pay reasonably well, even when they do not have to. One (Gneezy et al. 2010) also looks at PWYW combined with a share of proceeds to charity, which was found to be particularly effective (more in Chapter 22). Another (Regner and Barria 2009) focuses on a very interesting online indie music distributor with a PWYW model (combining a try-before-you-buy feature and a high revenue share with artists, both seen as enhancing payment levels). Natter and Kaufmann (2015) is one of the most comprehensive recent surveys of this body of work.
While many of these papers are written for academics, business people would do well to give them a look (skipping over the heavy parts, and with just reasonable caution that some of these models and experiments are simplified). Also, the Wikipedia article on PWYW is regularly updated and has useful information, including links to information on notable real uses of PWYW. In addition, I am collaborating on a paper with Adrian Payne and Pennie Frow that is to include an extensive survey of PWYW from the perspective of dynamically participative co-pricing strategies such as FairPay (to be cited at FPZLink when available).
Broad Grounding in Behavioral Economics
Nobel Prize winning economist Daniel Kahneman’s book, Thinking Fast and Slow (2011) presents a wide range of insights into how we think and make choices. These insights—mostly established in the past few decades, and only beginning to be widely recognized—form the groundwork for the new field of behavioral economics. Kahneman’s book shows how greatly people are influenced by framing and anchors. While our decision processes are wonderfully rich and nuanced, they are highly susceptible to influence. Much of our decision making is based on the fast, intuitive, but error-prone processes Kahneman calls System 1—with reluctant intervention by the slower, more deliberative and “rational” processes of System 2, which is often lazy, and also subject to systematic errors.
As we have seen, FairPay introduces a model in which the buyer chooses the price, but with incentives to consider whether the seller will judge it to be fair, in that particular context. Counterbalancing that, the seller has the ultimate power to motivate fair pricing by controlling whether future FairPay offers will be made to that buyer. The seller can also frame the offer, and set a suggested price as an anchor. Applying the principles of behavioral economics as outlined by Kahneman, the seller has many opportunities to nudge buyers to price fairly and even generously. Many of these lessons are very relevant to FairPay:
Framing is very powerful and applies to both System 1 and System 2. Choices can be altered very dramatically by framing the choice in different ways. The core objective is to make buyers feel that the seller deserves a fair price, and to find incentives for pricing generously.
Nudge (Thaler and Sunstein 2009) is the title of another book (based on work also described by Kahneman) that makes much of choice architectures that can nudge behavior in very powerful ways. A choice architecture is a systematic process for framing choices to induce desired behaviors. FairPay sellers can use such choice architectures to nudge most buyers to set prices to desirable levels.
Anchors are an aspect of framing, providing a reference point for a price or other parameter to be chosen. Even where the individual is free to ignore the anchor entirely, it tends to have a surprisingly strong influence on their choice. FairPay exploits the power of suggested prices as psychological anchors to nudge buyer pricing choices.
Intensity matching is one of the things the intuitive and automatic System 1 does easily. Matching prices to one’s happiness with a product is easy. Thus choosing prices that match to perceived value is also easy (see the next section on tipping), and the desire to pay fairly for being delighted by a seller can be harnessed just as well, and perhaps better, than the competing desire to find bargains.
WYSIATI (What You See Is All There Is)—people (especially their System 1), tend to make choices based on just the factors they see, and even System 2 is lazy and tends not to search for other factors that may be missing. A key objective of framing and choice architecture is to present those factors that cause buyers to notice what they should value (and to downplay any negatives) in order to nudge toward desirable pricing choices.
The laziness of System 2 is perhaps most strikingly evident in the use of opt-in versus opt-out choices to set desirable defaults. Thus making acceptance of a suggested price an opt-out choice (such as for ongoing subscriptions) might lead most buyers to be compliant, most of the time (especially once a comfort level is established).
Decisions are affected by distortions in psychological weighting.
Loss aversion is stronger than gain-seeking. Losses from an established position tend to be weighted more heavily than equivalent gains. Once a buyer has the privilege of setting prices under FairPay, they will be averse to losing that privilege (by pricing unfairly low just to get a small gain on one transaction).
Fear of disappointment also tends to be overweighted. The post-experience pricing of FairPay eliminates the risk of buyer remorse, and that fear can be a significant barrier to buyers accepting a price corresponding to full value up-front. Pre-set prices are implicitly discounted in the buyer’s mind to allow for that risk. FairPay removes that discount factor.
Perceptions of fairness and entitlement are inherently frame-specific and not underlying. By framing fair prices as being context-dependent—such as based on usage, value obtained, ability to pay, and other factors—buyers can be nudged to accept prices that are different from those for other buyers in different circumstances. Those who pay a higher price will not feel regret on learning that another paid less, if they are helped to see that the equity of the price depends on the circumstances (different entitlements in different contexts).
Building on how System 1 and System 2 partner in making choices, an objective of FairPay choice architecture is to make it easy for the buyer to accept seller-suggested prices so that both buyer and seller maintain a happy and mutually valuable commercial relationship.
Kahneman mentions cognitive ease as a key factor in how we make choices. When things are going well, with no surprises or threats, System 1 does its automatic control—but when there is cognitive strain, System 2 reluctantly jumps in. Obviously, FairPay pricing processes should seek to maximize cognitive ease, as they nudge buyers to fair pricing.
The causes of cognitive ease are:
Repeated experience: As buyers become familiar with FairPay, and with a given seller, they will gain comfort and ease.
Clear display: Offers and pricing requests should frame the offer and the suggested price as simply and clearly as possible.
Primed idea: Offers and pricing requests should frame the value proposition, the relevant context, and the suggested price to most effectively prime the buyer to accept it as fair.
Good mood: FairPay aligns the motives of the seller and the buyer—if the seller delights the buyer and positions himself as a positive and deserving partner and provider of value, the buyer will find it easy to reward that (just as most people give good tips to delightful and effective waiters, and pay premiums to companies that delight them—such as Apple).
The consequences of cognitive ease are: Feelings of familiarity, truth, goodness, and effortlessness.
All of this suggests that effective choice architectures can be applied to make FairPay very effective for most buyers—if the seller really seeks to deliver what the buyer values, in return for a fair profit for doing that.
Some aspects of choice architecture and nudging may smack of exploitation of buyers by manipulative businesses—but remember, this is in the context of a system in which buyers have complete power to set any price they think fair. The choice architecture is just a defensive tool for sellers to nudge buyers toward fairness in choices they have full freedom to make. The nudging of a choice architecture may be “manipulative” to a degree, but in the context of an unprecedented level of buyer freedom—sellers need some compensatory powers (and it seems that abuse is unlikely and easily countered).
Cognitive Ease and Lessons from Tipping
Further insight into the cognitive ease of FairPay can be seen from some parallels with restaurant tipping (more at FPZLink). Some people dislike tipping (especially outside the United States) but it has strong appeals. It gives patrons the right to match tips to the value of the service, rewarding good service (sometimes very generously) and pricing lower for poor service. It is widely recognized that when restaurant service is built into prices, with no discretionary reward for service quality, service is often poor and customers dissatisfied. There are of course some customers who tip unfairly. But the problem decreases among regular customers, and it is in just such long-term relationship contexts that FairPay seems most likely to do well (and use of FairPay can be limited to such contexts).
Tipping teaches us how easy it is for people to compute value intuitively. On hearing about FairPay, people often ask “isn’t it a cognitive burden for customers to have to think about the value?” But tipping shows that this kind of evaluation is not difficult—it is highly intuitive. We easily do a complex multivariate, multidimensional analysis in our head, during and after a meal, and know immediately whether we think the service was average, better, or worse, and by roughly what degree (Kahneman’s intensity matching). We can then easily figure whether to adjust our average tipping level up or down, plus whether to adjust for being a regular or any other special factors, to determine that we should tip X percent. (The only difficulty is just doing the arithmetic of how much X percent is—and that is easily automated away.)
So the behavioral economics of tipping is very supportive of the idea that FairPay will prove very effective in selected business contexts.
FairPay as a Repeated Game
Game theory offers very compelling support for FairPay as a repeated game, and perhaps can serve as a powerful tool to help design FairPay journeys that will push this game toward highly win–win behaviors. As noted earlier, the problem with most current pricing methods is that they are transaction-focused, as one-time games, and not relationship-focused, as repeated games. Structuring the customer relationship as a repeated game promises to add strong incentives to build a reputation for paying fairly—even at a short-term cost—in order to get a continuing stream of attractive offers in the future.
This seems a compelling argument that FairPay can be made to work very effectively once both sides understand the game. I have only limited understanding of game theory, but a paper on “The Evolution of Cooperation in Infinitely Repeated Games” (Bó and Fréchette 2011) seems to support that. The paper explains that there are a number of issues that affect the level of cooperation that is actually achieved, and that may not correspond to theoretical equilibrium, but “cooperation does prevail … when the probability of continuation and the payoff from cooperation are high enough.” That seems to reinforce my expectation that FairPay will generally work if the business makes the repeated game attractive to the customer, and applies reasonable (but not unduly harsh) controls on FairPay credit outstanding to limit the losses in cases where the customer fails to cooperate.