getting serious about changing behaviors

26 Apr, 2010 • posts i've written • 15 comments

“We’re in the business of changing customer behavior.”

Behavioral change has become the topic du jour in marketing, but as always, there’s little to show in way of actions per the volume of conversation. Moreover, it seems as though that the same tactics are being chased (a loyalty program through Foursquare anyone?) without a holistic strategy.

In this post, I’ll demonstrate how to use the Theory of Planned Behavior to develop cohesive strategies to encourage changed behaviors.

background

The Theory of Planned Behavior proposes a model for how human action is guided. It predicts whether or not a specific behavior will be taken, provided that the behavior is intentional, through developing an understanding of beliefs and attitudes. It was proposed by Dr. Icek Ajzen, a Professor of Psychology at the University of Massachusetts, and it is one of the most predictive persuasion theories (as shown in the application of study in advertising, public relations, and healthcare).

By using the theory, marketers and advertisers are better able to 1) understand the beliefs and attitudes of their customers, 2) segment their customers by beliefs, 3) create advertising or other interactions to impact those beliefs, and 4) measurably increase the likelihood of the desired behavior.

To predict whether a person intends to do something, we need to know:

• Whether the person is in favor of doing it (attitude)
• How much the person feels social pressure to do it (subjective norm)
• Whether the person feels in control of the action in question (perceived behavioral control)

If we maximize these three predictors, we will increase the likelihood that the person will intend to perform the desired action and thus increase the chance of the person actually doing it.

how to use the theory

Step 1: Define the Population of Interest

Who’s our target market? A word of caution: because using the theory involves surveying the audience, we should strive to choose a group whose size enables us to survey a representative sample. In other words, don’t say ‘everyone’ unless you’re able to pay to survey a representative sample of everyone. Examples include: moms in the northeast, teenagers in Texas, and senior citizens in Boca Raton.

Step 2: Define the Behavior

In order to use the Theory of Planned Behavior in our work, we must define the desired behavior. To do so, we should clearly state the behavior in terms of target, action, context, and time (TACT):

• Target (our product, our content, or whatever the behavior is to be acted upon)
• Action (purchase, download, log-in, the act we want our customer to take)
• Context (in the gym, at their computer, in the store, the environment of the action)
• Time (on Mondays, in the morning, every day, on every visit)

We can be as specific or as general as we’d like about the behavior using TACT, but we should ensure that the level of specificity is maintained throughout our process. For example, if we asked consumers the questions ‘Would you purchase an SUV?’ and ‘Would you purchase an SUV from your local Ford dealer this fall?’ – each question could undoubtedly result in a far different response and cloud our overall results.

Step 3: Survey the Population

The next step, after we’ve clearly defined our desired behavior, is to develop an understanding of our population’s intentions. As the graphic above noted, an intent is formed through attitudes, subjective norms, and perceived behavioral control. By administering a survey of our population, we’ll be able to measurably predict the likelihood of action and begin to explore how to maximize that likelihood.

Attitudes are assumed to have two components: beliefs about the consequences of the behavior (e.g. quitting smoking will decrease my risk for lung cancer) and the corresponding positive or negative judgements about the consequences of the behavior (e.g. decreasing my risk for lung cancer is desirable).

Subjective Norms are the individual’s own estimation of the social pressure to perform or not to perform the desired behavior. Subjective norms are also believed to have two components: beliefs about how other people, who may be important to the individual, would like them to behave (e.g. I feel pressure from my wife to quit smoking) and the corresponding positive or negative judgements about the consequences of the belief (e.g. in the context of quitting smoking, doing what my wife thinks I should do is important).

Perceived Behavioral Control is the extent to which a person feels able to enact the behavior. One part of this is how much a person has control over the behavior (e.g. high control over putting a cigarette either in my mouth or not) and the other part is measured by how confident the person is in performing the behavior (e.g. low confidence due to nicotine addiction).

Building the survey – I suggest consulting Dr. Ajzen’s paper on the topic before you administer a survey of these predictors (he’s also made available a sample questionnaire for review).

Step 4: Analyze the Results

Once we’ve surveyed a representative sample of our target population, we’ll have a thorough understanding of their collective attitudes, subjective norms, and perceived behavioral control.

In addition to drawing high-level conclusions (e.g. our population feels in control but doesn’t possess a strong enough attitude to change behaviors), we can use our data to further segment our population: one group could consist of members that don’t have adequate beliefs about the desired behavior (e.g. those that don’t believe that smoking causes lung cancer) or a group could consist of members that feel strong social pressures to perform the behavior (e.g. those with families strongly advocating that they quit smoking).

Step 5: Design Strategies to Maximize the Predictors

Once we have an understanding of our three predictors (attitudes, subjective norms, and perceived behavioral control), and their areas for improvement, we can begin to design strategies to measurably impact those deficiencies (or take advantage of strong areas).

While we ideate on strategies and tactics, it’s important to:

• create an ecosystem of solutions to maximize all three predictors among our population and thus maximize the likelihood of the desired behavior
• pay special attention to the implementation of strategies and how different segments of the population react differently (e.g. this recent example of behavioral economic principles applied to a real world problem showed how different audiences with different beliefs have very different reactions to a strategy, in this case reinforcing the negative behavior)

Example strategy: Imagine that after we’ve analyzed the results of our survey (let’s imagine we’re testing for the behavior of quitting smoking), we’ve found that our population believes that smoking is harmful, wishes to quit, feels pressure from their family & friends to quit, desires to satisfy them, but feels it’s too difficult to quit due to nicotine addiction (control over the behavior). Our strategy (as has already occurred), would be to help people feel that they have more control over their nicotine addiction. From there, we could ideate all manners of pharmaceuticals, technology, and advertisements to make it easier to reduce the addiction (increasing the perceived behavioral control) and to communicate those benefits.

Step 6: Re-test and Refine

Your results have an expiration date, especially if you’re actively trying to influence beliefs and attitudes. Revisit your population, re-administer the survey, and refine your strategies based on these findings in a timely manner like any good scientist would.

examples to consider

I thought it would be helpful if I chose a few examples of marketing or advertising and explained how they could be impacting our three predictors to influence behavioral change.

Your Brain On Drugs is a classic example of an advertisement targeting attitudes. This commercial is attempting to reinforce the consequence of a behavior (taking drugs) and a negative judgement of that behavior.

Friends Don’t Let Friends Drive Drunk is a great example of an advertisement that is trying to increase social pressure (subjective norms) and the need to satisfy that pressure.

Of course, these are both heavy handed attempts to maximize a predictor and they’re truly traditional forms of advertising in themselves (the PSA), but it is quite possible and perhaps more effective to be less overt and smack less of traditional advertising.

Nike Plus, in comparison to the previous examples, was/is much more than advertising alone, it’s an interactive platform to reach people and reinforce a changed behavior (running more often, therefore needing new shoes more often). The system offers feedback, recording how often you run and how well you run, which supports your perceived behavioral control (you can run, because you have run). You can also share your data with friends in the community which can create further social pressure (subjective norm) to keep up with the behavior. Moreover, it doesn’t hurt to watch someone you know with a Nike Plus gradually lose weight which bolsters your belief that running more often is good for you. All in all, Nike Plus is an extremely well crafted tactic devised to influence changed behaviors.

I hope you enjoyed the post. It took quite a bit of time to create, so if you did, please do share it and/or leave a comment. To influence your behavior, I’ll remind you that substantive comments make you seem intelligent to others, sharing the post on Twitter makes me like you all the more, and if you scroll to the bottom of the post you’ll see a Twitter widget on the right which makes sharing the post incredibly easy.

further reading

If I’ve piqued your interest on the topic, I highly suggest the following sources:

• Dr. Ajzen’s personal site with more information on the Theory of Planned Behavior
• His full paper on the topic from 1991 [PDF]
• His new book, Predicting and Changing Behavior