Three things to consider before crafting a behavioral intervention

We are continuing our exploration into behavioral science and its practical application to ethics and compliance. In previous posts, we’ve talked about behavioral risk hot spots, how to identify them and what’s the role of the underlying behavioral drivers. The next step is to mitigate the identified risks by changing behaviors.

There are three typical steps of behavioral change:

• understanding the relevant behavioral risks and key focus areas,
• formulating very specific desired behaviors, and 
• setting up the interventions.

But wait, is there anything critical to know upfront before crafting an intervention? Any specific requirements to satisfy or resources to put in place? Yes, absolutely. Here are three key things to consider.

Start small

Human behavior is very complex – many factors influence our decisions, big and small. Now think about an average organization. It is a super-complex non-linear system, a massive interplay between various group norms, behaviors, and underlying drivers with a rarely occurring single cause and effect. 

Acknowledge upfront that you don’t know much about it, be all eyes and ears and start by making small changes. Then see which impact it has on the actual behavior – apropos, sometimes this may not be how you intended. Be careful with “good” norms and behaviors that you want to keep intact – they may disappear or change because of your intervention. So be wary and take baby steps. 

Some behavioral interventions, especially those focusing on i-level change like nudges, may seem simplistic, but this is fine. For example, does changing parts of an email impact behavior at all? Apparently, it does. In the longer-term influencing people’s choices, even small ones, can have a bigger cumulative effect on the organizational decisions and, ultimately, the outcomes. And that’s what you’re aiming for.

Explore and experiment

First, working with behaviors is an experimental, trial-and-error approach with no point-and-click solutions. Some interventions might not work – either because they don’t change the targeted behavior at all or because they change it not in a way you intended.  

Second, behavioral risk is very context specific. It means that what works for one team or location might not work for another. Replications and extrapolations may be ineffective, if not dangerous. Subculture focus is key.

Third, measuring the effectiveness of interventions can be tricky because when you start measuring things, you start manipulating them. And that, in turn, may also change the behavior.

You might think – who on earth will give me a license to “experiment” with all that? And that brings us to the next point.

Have a senior sponsor

This is crucial. Find someone from the leadership team who will be on board with experimenting. In the banking sector, where the behavioral approach recently gained a lot of traction, it is usually a Chief Risk Officer or a Chief Audit Executive.

You could use a couple of selling arguments to get these people’s buy-in. First, experimenting with interventions is innovative and forward-looking. Second, unlike other innovations, the incremental cost is limited – you won’t need to spend a fortune to try it. Third, the results may surprise you disproportionately. Some behavioral practitioners have already tried this narrative – and say it works well with the boards.