When is behavioral influence desirable and when is it not?

When you visit a Web site these days, chances are you are a participant in multiple online experiments. Experiments that examine what modifications to a Web site's choice structure change the behavior of the people who visit the Web site.

For these experiments, you, the participant, are not asked for conscious consent. And yet, within scientific research there are strict requirements to protect the participant.

But how problematic do people actually find it when they are subjects in one or more experiments during their visit to a website? And how do we feel about the insights gained from these experiments being used to influence our behavior?

Choice architecture is inevitable

Before I elaborate on the above question, it is important to reflect on the fact that, to a certain extent, a choice architecture (the way choices are presented to us) and thus influencing our behavior is simply inevitable.

For example, every store has a layout where certain products are seen before others, every menu ranks different dishes, and similarly, every website has a design that influences what people see and thus, how people make choices.

In short, online or offline influencing is found everywhere and is inevitable. But there is a difference between ‘just influencing’ and ‘targeted influencing based on insights gained from behavioral research.

People want to be able to make a conscious rational choice

A fair amount of research has been done on people's opinions about encouraging a certain type of behavior through nudges (small nudges in the right direction). These studies have shown that people generally prefer influence that is directed toward our deliberate and conscious decision-making processes. Consider interventions that provide information about the best course of action. Such as adding nutritional information on food packaging.

Influences that target our automatic, non-deliberative decision-making processes, on the other hand, are more likely to be frowned upon. These are mainly interventions that respond to our human biases (fallacies) such as catering to the need to make the same choice as others have done for you (conformism). Such forms of influence make people feel they are losing their freedom of choice and autonomy.

Preferences regarding nudges are malleable

However, follow-up research shows that this distinction is not nearly as black and white and that people's preferences toward nudging depend on the situation and how you ask the question. For example, people seem to have no problem with behavioral influence when they agree with the values of the people implementing the interventions and/or when the goal is in line with their interests, political opinions or their values.

However, there does seem to be a general negative attitude when the interventions cause economic or other losses by capitalizing on people's inertia, inattention or other common thinking errors. In other words, when nudges in a commercial setting capitalize on our automatic, non-deliberative decision-making processes in order to increase sales.

Online influencing is seen as negative

I also see the same patterns recurring in my work. For example, a while ago a fellow psychologist gave a presentation in which she used a number of statements to try to find out how people viewed online experiments and online behavioral influence in general.

When she started with the statements, it appeared that people strongly disliked online behavioral influencing because they assumed that the organizations would not have their best interests at heart. For example, when she asked if the same behavioral influence was desired offline, they answered, “yes, because a supermarket just wants to encourage them to eat healthier.”.

Our opinion differs from our experience

So when you ask people, they seem to have a negative view of experimentation and the influence that results from it mainly in a commercial setting that takes place online. Let this be precisely the sector where these practices are increasingly being carried out.

But as I said, this is the answer when you ask people for their opinions. As psychologists, we know all too well that the opinion we think we have often does not match our opinion in reality. In practice, people often don't even realize the influence, let alone actively protest it.

However, this does not mean that we can be pushed in all directions without boundaries without realizing it and blindly accept all forms of influence. All elements in our environment consciously or unconsciously influence our behavior and experience. So despite a person being far from aware of the complete choice architecture or direct influence on a site, it is experienced. For example, a strongly guided choice will evoke a negative feeling in your customer.

When is behavioral influence desirable and when is it not?

As an organization, you want the customer to have an overall pleasant experience, otherwise it will simply have negative consequences for your growth. This means a good balance of influence that on the one hand gives visitors enough information to make an informed choice, but on the other hand provides emotional support. For example, when the visitor is unable to make a fully reasoned and informed rational choice.

Exactly where that balance lies depends on the organization, the type of product and the type of customer. By experimenting, you will learn step by step what works for your customer and what doesn't. Just be sure to properly determine when the effect of a particular influencer is the right one. Common metrics such as click through rate or even the increase in conversion can fall short here because they only tell you something about the short-term effect.

Your customer's experience and the feeling the customer gets from that experience transcends the visit to your Web site. In other words, a conversion is not a direct reflection of a satisfied customer. Therefore, always make sure you know what the long-term effects of your influencing are and think carefully about how you determine when the effect is desirable (i.e., whether the effect contributes to your customer's well-being) and when it is not.

Research on the long-term effect of nudging

In collaboration with the Vrije Universiteit Amsterdam, Professor Kobe Millet and Peeter Verlegh, my colleague Eline van Baal has been busy setting up a PhD. The goal of this PhD is to validate scientific behavioral insights and also to develop a method to make long-term behavior change techniques more measurable. Read here more about the study and how you can contribute.

Literature:
Joffe, S., Cook, E. F., Cleary, P. D., Clark, J. W., & Weeks, J. C. (2001). Quality of informed consent: a new measure of understanding among research subjects. Journal of the National Cancer Institute, 93(2), 139-147.

Schultz, D. P. (1969). The human subject in psychological research. Psychological bulletin, 72(3), 214.

DAVIDAI, S., & SHAFIR, E. (2018). Are ‘nudges’ getting a fair shot? Joint versus separate evaluation. Behavioural Public Policy, 1-19.

Sunstein, C. R. (2016). Do people like nudges? Administrative Law Review, Forthcoming.