Changing the key behavior
REV is a dynamic pricing tool that automatically sets rates for linear advertising based on fill rates and optimizes spot placement.
Even in today’s high-tech world, price setting in radio broadcast advertising is still very manual. Created a few times a year, rate cards rarely reflect pricing based on current inventory, and the sales rep is left to manually count inventory and estimate the true rate. REV is a dynamic pricing tool that takes these manual steps out, but sales reps were still reverting to their old ways and not using the tool the way we intended.
Turning to insights from behavioral economics, I created several experiments grounded in behavioral science that aimed at changing a key behavior: Get radio advertising sales reps to leverage the spot optimizer the first time they create an order.
Cognitive Bias Analysis
A Behavioral Science Approach
Behavioral economics says that people predictably act unpredictably. Because of this, it’s unreliable to ask a user if they would use a hypothetical solution or ask them what needs to be changed for them to adopt it.
According to behavioral science, users are mostly unaware of their cognitive biases and, therefore, the environment needs to be designed to offset these biases.
I used a 3B analysis framework to approach this particular problem:
- Identify the desired key behavior, then outline every step customers have to take to achieve it.
- Reduce the barriers that get in the way of decisions.
- Amplify the motivating benefits.
Desired Key Behavior
Identifying a key behavior helps scope the analysis toward a crystal clear action that the user should take in order to impact the product outcome.
REV aimed to help users optimize revenue (product outcome), but in order to do that, salespeople using the sales software needed to do things a little differently than they were used to.
So, our desired key behavior was to get broadcast sales reps to leverage the Spot Optimizer every time they created an order.
A behavioral diagnosis starts with a map of each step a user takes toward the desired key behavior. It’s not an aspirational map, but a representation of the user’s current journey. Even a decision is mapped out as a step.
The next stage involved identifying barriers in the user’s behavioral map.
You can see that several cognitive biases created barriers.
After identifying what we thought might be barriers for users executing on our desired key behavior, I conducted some quick usability tests to validate these assumptions. Were users really just not seeing the optimizer (attention bias)?
While that was true – it was almost unrecognized on the page – what the usability tests also revealed was an even more complicated bias created by the ambiguity effect. People didn’t fully understand the benefits of the optimizer, so they defaulted to what they knew even though the other option, manually selecting advertising spots, was time-consuming and prone to user error.
Intervention Design & Mock-ups
With a behavioral science twist, I came up with ideas for reducing barriers before turning them into wireframes. These solutions were added to the behavioral map as interventions.
Since we were looking to see a change in our user’s current behavior, we mocked up the UI in order to mimic the environment they are normally used to seeing.
Without a sizeable user base, it didn’t make sense to roll out a live experiment on the product itself, although these kinds of experiments are best executed to larger sample sizes with experimental conditions and a control.
So we conducted more user testing with salespeople and with a varying degree of experience selling in the radio broadcast industry. The user tests revealed more information about our target users’ motivations and we are making adjustments based on those learnings.