Creating A New Safety & Justice Playbook

 

FrameWorks Institute

An innovative pilot explores the power of persuasion testing & digital ads to shift sentiment & influence narratives

 

The Challenge

Public sentiment around crime and safety can pivot on a dime. The issues are deeply personal and emotional, making them highly prone to weaponization. This dynamic emerged in full force in the post-pandemic era. A narrative of diminishing public safety took hold in cities and counties across the country, blaming criminal justice reform broadly for increases in crime, violence, and disorder, whether real or perceived. 

Enter FrameWorks Institute, 1235 Strategies, and Daigneault Digital, who surfaced two critical yet unanswered narrative campaign questions: 

  • At the local level, where so many issues of crime, safety, and reform play out, how can groups on the ground use available opinion research to reframe changes to local criminal justice systems so they are understood as safety solutions—not problems? 

  • And how can local groups start reaching and engaging moveable audiences with persuasive messages upstream of narratives forming in the press and other organic channels?  

 

our Solution

With support from the Safety and Justice Challenge at the John D. and Catherine T. MacArthur Foundation, and in collaboration with Yosemite Consulting and several* other seasoned campaign strategists, we launched a pilot in late 2022 to answer these narrative questions.

Our approach centered on two things: 

  1. Combine available opinion research with randomized control trial (RCT) message testing to identify the frames, messages, and creative that persuade moveable audiences in two distinct communities—St. Louis County, MO, and Multnomah County, OR—to support local reforms and other non-carceral outcomes identified and prioritized by groups on the ground. 

  2. Apply a page from the political playbook—an emphasis on message repetition and saturation through digital ads—to a 501(c)(3) narrative setting to test whether moveable audiences in St. Louis and Multnomah counties would watch and engage with persuasive creative about why local justice reforms are safety solutions—not problems.

Ultimately, from July 2023 to July 2024, we executed 30 message and content tests, conducted two audience surveys and 20 in-depth interviews, and ran two four-month ad campaigns.

 

the Results

In St. Louis County, where press stories about increasing crime rates and tragic incidents had influenced the local public safety narrative, the most persuasive message was “Break the cycle”:

“Did you know that we put people in jail who are not a threat to public safety and before they're put on trial? They sit there for weeks, months, even years, and the things that drive crime get worse, like lost housing and jobs or mental health and addiction. When they leave jail, they're more likely to be rearrested, and the cycle of crime begins. But we can stop this from even starting. Instead, address the things that contributed to their arrest with mental health and drug treatment and other solutions. This makes them less likely to be arrested again and makes us safer. St. Louis County, let's break the cycle.”

In RCT message testing on Swayable, where the benchmark movement achieved in advocacy message testing is 0.7 point, this broad message significantly moved multiple audiences to support an investment in diversion programs by 3.0 to 5.2 points.

We distilled the “Break the cycle” message into several ads, with “We all want” proving to be the most persuasive:

Based on RCT message testing on Swayable, this winning ad:

  • Moved all audiences to support speedier post-arrest access to an attorney and increased investment in diversion

  • Moved people 55+ and independent white women to support all policy outcomes identified as priorities by local partners (i.e., speedier access to an attorney, diversion, bail reform, and a prevention-first approach over tough-on-crime) 

  • Moved white, postgrad, and very conservative audiences—as well as registered Republicans—to support bail reform.  

  • Generated huge movement for white Democrats 55+ to support a prevention-first over a tough-on-crime approach

A second ad in St. Louis County—“Better”—moved all audiences to support a prevention-first approach over tough-on-crime and was especially effective with registered Republicans. (By contrast, “We all want” moved only people 55+ and independent white women to support prevention-first.) As a bonus, “Better” also moved liberal and white Democrats 55+ to support bail reform.

In Multnomah County, where issues of crime and safety were weaponized around several high-profile moments for residents (e.g., a DA election) and decision-makers (e.g., maintaining or repealing Oregon’s drug decriminalization law), the most persuasive message was “Treatment, not jails”:

“Rather than throwing people in jail where they can't get effective help, we need to build a better system—one that connects people with services and treatment in settings that actually work.”

This much narrower message significantly moved multiple Multnomah County audiences by 4.2 to 4.8 points to support drug treatment on demand, a part of Oregon’s drug decriminalization law that groups on the ground hoped to maintain (again, the benchmark movement for advocacy messages on Swayable testing is 0.7 point).

We distilled the “Treatment, not jails” message into multiple ads, with “Recipe for change” proving to be the most persuasive:

Based on RCT message testing on Swayable, this ad increased support across all audiences on diversion and providing a continuum of care to address addiction, another part of Oregon’s drug decriminalization law that local groups wanted to preserve. It was especially effective with:

  • Urban audiences

  • People 55+

  • People making less than $40K

  • People who leaned more conservative 

  • People who regretted voting for Measure 110

A second ad in Multnomah County—“Monta”—featuring a personal testimonial from a local ally, was especially effective in moving suburban audiences to support providing a full continuum of care for people with addiction. As a bonus, it also moved conservatives to support a prevention-first over a tough-on-crime approach.

Finally, from Swayable testing, we knew that when audiences in St. Louis and Multnomah counties watched our full 30-second ads, we persuaded many of them to support changes to their local criminal justice systems. But would people watch the ads in their entirety when we put them into the market and they showed up on their devices? 

In both counties, despite most of the video ads being skippable, we generated a 69% video completion rate, beating the industry's 60% benchmark.  

We also tracked social media engagement rates and saw rates on our ads were four to six times higher than industry benchmarks. While engagement rate wasn’t our key metric, it served as another proof point of issue salience and message resonance. 

 

Lessons Learned

Based on insights from RCT message testing and the digital ad campaigns running in two distinct counties, we created the Safety and Justice Playbook: Shaping local narratives through persuasion testing and digital ads.  

The playbook offers five key ideas to help groups on the ground, their funders, and their allies move opinion research toward narrative change execution. Each key idea is backed by examples, test results, and takeaways from the pilot. 

  • Key Idea #1: Frame changes to local criminal justice systems as safety solutions—not problems—by taking a pragmatic stance on doing what works to improve the thing we all want (safety) and not doing the things we know don’t work. 

  • Key Idea #2: Prioritize the strategic use of digital ads to shift sentiment about local criminal justice reforms and related non-carceral outcomes upstream of narratives that appear in the press. Apply Key Idea #1—taking a pragmatic stance on doing what works to improve the thing we all want (safety)—to inform creative.  

  • Key Idea #3: In local efforts to shape narratives among moveable audiences, disregard preconceived assumptions about who is moveable. Far more people are persuadable about changes to local criminal justice systems than we realize, so begin your effort by counting people in—not out. Prioritize framing, messaging, and creative that are persuasive across demographics, and treat audience targeting decisions for execution separately. 

  • Key Idea #4: Consider potential messengers who are most likely to strike the pragmatic stance in Key Idea #1: doing what works to improve the thing we all want (safety) while reflecting local priorities and contexts. 

  • Key Idea #5: Start playing the long game—consistently and proactively communicating persuasive messages around changes to local criminal justice systems. If your offensive posture is a pragmatic stance on doing what works to improve the thing we all want (safety), you’re in a better position to address threats to your priority reforms and outcomes head-on when they arise. 

 

*We’d like to thank our additional pilot partners, including 76 Words, Nikc Miller, HonorCode Creative, and Change Research.