A Better Way to Provide Precommitment Offers: First “Now,” Then “Later”

Journal of Marketing Research Scholarly Insights are produced in partnership with the AMA Doctoral Students SIG – a shared interest network for Marketing PhD students across the world.

Consumers make numerous decisions every day, with consequences in the near or distant future. Some decisions that result in farsighted behaviors—decisions involving immediate costs but offering delayed benefits—can be critical in promoting consumer well-being. However, consumers face challenges when attempting to adopt such farsighted behaviors, primarily because the immediate costs loom larger and overshadow any prospective benefits. For example, deciding whether and when to start saving for retirement involves many decisions that can act as barriers and prevent consumers from enrolling in an investment plan to achieve financial security.

To counteract this, marketers commonly use precommitment strategies by inviting consumers to choose between adopting the farsighted behavior immediately or in the future (i.e., now vs. later). Providing the option to “do it later” can make the decision feel less costly, thus reducing the aversive feelings consumers experience and nudging them to partake in farsighted behaviors. If this conventional wisdom holds, one would expect precommitment strategies to consistently facilitate the adoption of farsighted behaviors. However, in a recent Journal of Marketing Research article, researchers find that these strategies can sometimes backfire, depending on how marketers display precommitment choice options to consumers.

Design and Urgency in Consumer Choices

The researchers identify the role of urgency as a novel explanation for why certain types of precommitment strategies can fail to produce intended results: The temporal order of the choice display influences the implicit level of urgency that a marketer signals to consumers to engage in farsighted behavior. Specifically, when precommitment options to save now versus later for retirement are displayed simultaneously, or side-by-side, it can coincidentally reduce adoption of the behavior change (i.e., saving) because consumers infer that the way in which the choices are presented signals that it isn’t urgent for them to engage in these behaviors. In other words, when both options are visible to consumers at the same time, they are less likely to engage in farsighted behavior because they infer the marketer is signaling that it isn’t urgent to do so.

When both options are visible to consumers at the same time, they are less likely to engage in farsighted behavior because they infer the marketer is signaling that it isn’t urgent to do so.

In contrast, sequential precommitment options that display a single option to adopt the behavior now, followed by the option to adopt the behavior later only if the consumer initially declines, signals a greater sense of urgency to encourage consumers to adopt the farsighted behaviors. The researchers conducted a multisite field experiment and two lab studies, finding that the sequential precommitment strategy is more effective than using a simultaneous precommitment strategy or no precommitment strategy at all.

When designing choice sets to encourage precommitment, marketers and policymakers are often tempted to provide the option for consumers to choose to “do it later,” as doing so can reduce the psychological burden consumers bear in the present. However, this research suggests that such a strategy could fail because people may feel that it is not urgent to engage in this behavior. Offering the option to “do it now” and then “do it later” only if consumers decline can boost the effectiveness of precommitment offers and, ultimately, nudge consumers along the continuum to adopt vital behaviors that can contribute to their prosperity and well-being. When implementing interventions, practitioners should think through what might be signaled by the intervention, as the way to communicate offers can ultimately impact consumers’ decision making.

We go behind the scenes with two of the authors to learn more about their perspectives on the benefits of collaborating with practitioners to solve real problems in the marketplace, the surprising result that inspired their theory development, and to receive advice on the importance of crafting a message that sticks.

Q: Can you talk about some of the benefits and challenges that researchers and practitioners collaborating to conduct research may encounter when embarking on designing and conducting a field experiment?

A: We think a major challenge for academic researchers and practitioners collaborating on field experiments is that there can be potentially competing sets of objectives that aren’t completely aligned. So, the researchers are interested in conducting a study that is linked to theory and that somehow informs theory through testing hypotheses. The practitioners find that perfectly interesting, but they also want to make sure that the research project is something that helps them deliver the results that are a key part of their business and broadly learn how they can be the best marketers that they can be in their particular context. Those perspectives don’t necessarily conflict, but sometimes they do. For example, sometimes, a researcher’s hypothesis might be best tested with a variation on the status quo that is less effective than what is being done in practice. You don’t want to intentionally subject your wonderful practitioner-collaborator to that. Although this is initially framed as a challenge, it can also be a very important benefit because it really disciplines us as academic researchers to make sure that what we’re doing has direct practical implications.

Q: Can you elaborate on why you incorporated an incentive compatible design in Study 3 and why you thought it was important to align the participants’ incentives with the design of the intervention and marketplace goals in general?

A: We anticipated that it would be difficult for consumers to accurately simulate their actual behavior in the context that we studied. For farsighted decisions—like enrolling in a retirement plan, getting a flu vaccine, or updating your antivirus software in a prompt way, as opposed to putting it off for weeks and months—there are competing ways that people think about their options. For example, “In this moment, right now, what is going to be the easiest? What is going to be the most expedient? What is going to be the most immediately rewarding with the least upfront cost?” That’s what I want to do in this visceral way right now. As opposed to: “What I should do if I were to step back and reflect, and think about the all-in costs and benefits integrated from now over the infinite horizon, probabilistically weighted over how long I’m going to survive with some appropriate but mild rate of intertemporal preference.”

Because we know that there are these two competing ways of approaching these sorts of decisions when we’re asking about it in a scenario study, we don’t have a lot of confidence regarding which of those approaches participants are going to be accessing when they’re responding to the scenario that we put them in. We’re interested in the ultimate impact of our types of precommitment on decisions that people take, and we wanted more confidence that the objectives and motivations that we’re ascribing to participants are what are driving their decisions when they’re generating data in our studies. Because it’s incentive compatible, we’re capturing whichever mode of decision making is actually driving the behavior, which we felt was particularly important in our context because it allows us to test our full theory while looking at real decisions with real stakes.

Q: What do you hope to see changed in the marketplace because of your research?

A: As a direct implication, when marketers offer precommitment, we think it’s likely going to lead to higher enrollment rates if they choose a sequential design rather than a simultaneous design.

More broadly, when marketers or choice architects design interventions, they might implicitly signal some information about their beliefs and motivations, and consumers use that implicit information when making their decisions. For example, even if the marketers didn’t realize it when they were designing precommitments by offering simultaneous options side by side, it signaled a lack of urgency. We think this applies to other interventions as well.

So, in the design process, when trying to take an idea from behavioral science or psychology and integrate it into a field setting to produce positive behavioral change, it’s important to first audit that intervention and think through the potential negative inferences or negative pieces of information that might be signaled by this intervention, then redesign to try to guard against those potential negative inferences before running the intervention across a large group of people in the field. Auditing before scaling is something that our paper advocates for, and we think it can be very useful in applied behavioral science.

Q: Were there any “behind the scenes” stories that do not appear in the paper that might be interesting to share with us? Any surprising or “aha” moments?

A: When we were talking with the record keeper that implemented our field experiment to build on the idea of precommitment that Shlomo Benartzi had pioneered with Richard Thaler in their paper “Save More Tomorrow,” we thought we could leverage behavioral insights to help people save. Specifically, pushing things off into the future takes advantage of present bias. Those upfront costs for saving—the sacrifice of current consumption—that loom large are now pushed off into the future, making them less onerous. In the original implementations of “Save More Tomorrow,” in meetings with their clients, financial advisors first offered the option to save now, and then only if their clients declined to save now, the advisors offered precommitment. What became very clear is that the sort of communication occurring between an advisor and potential plan participants was something that wouldn’t as easily scale in the settings that we were looking at because it’s costly for a plan advisor to sit down with individuals. So, we were thinking, “Why don’t we scale up the idea by making it much lower cost? We’re going to do it over mailers.” Here’s where this was a blind spot of ours: We thought, “We can’t do this back and forth consisting of interactions between the advisor and the investor when we’re just sending out a letter. Why don’t we just collapse it all into what we now call “simultaneous precommitment.” We’ll offer you the now option and the later option to increase your savings rate, all on the same page.”

That was a situation where our research ambition met with the reality of what a record keeper and these plan sponsors could do, and we inadvertently changed the psychology of the precommitment design that we thought we were going to study. We think it turned out much more exciting this way. What we now call simultaneous precommitment led to less savings because it caused people to start saving later. That’s when we concentrated on figuring out our best understanding of why this might be going on.

Q: What kind of strategies did you use to effectively communicate your focal message and primary findings? Do you have any tips for practitioners or junior scholars who want to refine their communication tactics?

A: All the authors were really focused on the fact that we were communicating something that was potentially complex. We spent a lot of time trying to reduce complexity and make it as understandable as possible. Making the abstract concrete is probably the best strategy. Although “sequential” and “simultaneous” are nice words and accurate words, sometimes a picture is the best way for people to see the idea. The visuals we added to the paper really helped to distill our ideas into something tangible for readers (i.e., see Figure 1 on page 1096). Another strategy is to continue to revise until your work feels like something that someone who has never even thought about your topic before can pick up and easily understand. If something seems complex to us, it’s going to be even more complicated for the readers who are unfamiliar with the topic.