There Is Science (and Power) Behind Effective Citizen Communication

SCOTTSDALE, Ariz. — In local government, sharing information with the public can seem straightforward. Many cities have a range of communication tools—websites, newsletters, social media networks and media teams—as well as newspapers or television stations that cover city news.

But determining whether these messages have any influence on the public is another matter. Is the message being received? Are people doing anything as a result?

These questions vexed us in the city of Scottsdale, Arizona, as we sought to communicate with our 242,500 residents. We realized that many of our public messages about local government programs and services were often shared without testing whether they really achieved the desired effect.

When the city was invited to join Bloomberg Philanthropies’ What Works Cities initiative in June 2016, a unique opportunity arose to team up with WWC’s partner the Behavioral Insights Team to test whether our messaging about city services and programs could be improved.

Specifically, could people be influenced to take a specific action through more thoughtful and intentional message design?

Increasing Donations for Local Nonprofits

Our first project was an effort to boost contributions to our Scottsdale Cares program, which encourages city utility customers to donate $1 per month to help local nonprofits in our community. We used emails to utility customers to test whether offering an active choice would increase the likelihood of donations. Scottsdale residents were randomized to receive either no email, an email with a button saying “Yes” that was hyperlinked to the signup form, or an email with both a “Yes” and a “No” button.

Recent studies have shown that presenting an active choice can lead to engagement with a service. Even though residents are not compelled to make any choice at all, by presenting a noticeable “No” button in addition to the “Yes” button, we aimed to induce more active consideration of an activity and, in turn, generate more donations.

There was initial trepidation among our staff about giving customers an option to say “no”—but we decided to trust our partners at BIT and test the option to see what worked. Success was defined by whether a donation was made within 30 days of the email being sent out.

The results showed that giving people a chance to decide for themselves improved the contribution rate. Using data on email open rates, we calculated that the donation rate for that cohort more than doubled among customers who received the email with only a “Yes” button and more than tripled for customers who received the email with both “Yes” and “No” buttons.

Clearly, providing an active choice made a difference for this message.

Getting More Customers to Go Paperless

For another project, we wanted to encourage customers to sign up for paperless utility billing. Sending out paper bills costs us about 25 cents each month per customer. With 56,328 paper bills still going out, we knew that converting people to paperless billing would result in significant savings.

We tested whether a utility bill insert could influence customers to sign up for paperless billing. About half of these customers were randomly assigned to receive the business-as-usual bill, and the other half received the insert.

The insert was designed using the concept of social norms, emphasizing that “thousands of Scottsdale residents” had already gone paperless. Social norms are a powerful insight for encouraging the adoption of specific behaviors because we are highly influenced by what other people do.

We found that people who received the insert were 56 percent more likely to go paperless, saving around $6 per person who signed up. If brought to scale, this tactic could save nearly $200,000. Cost savings in the utility area help reduce future cost increases. When drilling into the data, we realized that change was primarily driven by customers who were already paying online but still receiving paper bills.

What Your City Can Do

Based on our experience with these and other similar tests, here are some tips if you are considering using behavioral insights in your city:

• Have a tangible and impactful goal, and a clear short-term outcome—things like increasing donations to human services programs, increasing diversity in police recruitment, increasing participation in employee retirement savings programs, and encouraging people to switch to paperless bills.

• Identify and gain access to real and relevant data to be able to demonstrate the effects of your tests, and ensure you have a substantial sample size, typically 800 to 1,000. For example, we obtained a list of 6,500 community college students to test police recruitment messages, tested 25,000 utility emails to increase donations to nonprofits, and 56,000 utility bills to encourage uptake of paperless billing.

• It’s not just about communicating better—although that does help. Sometimes people think that these efforts are just marketing gimmicks, but it’s much more than that. The trials are based on prior research in the fields of behavioral psychology and behavioral economics, and they can yield meaningful results for your city.

What We’ve Learned

These tools don’t fix all our problems, but they do help us figure out where we need to focus our efforts. Our Scottsdale Cares trial showed us that we needed to make donating easier, and we needed to clarify what the donations are used for. These insights are leading us to adjust messages and tactics in other areas.

After starting down this road, we realized we needed to find and develop the skills of people in City Hall passionate about using behavioral insights and testing what works. One such employee has coordinated two trials, is actively advocating that other programs and services undertake these efforts, and is now leading our team that is using data to identify new areas to test. We look forward to learning more from the results.