Using Behavioral Insights to Market a Workplace Safety Program: Evidence From a Multi-Armed Experiment

This Abt Global paper describes how behavioral marketing components added to an Occupational Safety and Health Administration (OSHA) brochure failed to increase demand for OSHA’s On-site Consultation safety consultation service.

Abt used a randomized experiment to test the effectiveness of several versions of a marketing strategy based on behavior change concepts. The strategy uses low-cost “nudges” to try to influence individuals’ behavior and decisions by removing seemingly minor obstacles to setting up consultations. One example: OSHA’s mailed brochures provided a state’s OSHA web address and phone number for employers to visit or call to schedule a consultation. We made it a little easier by sending emails with links to the scheduling website so that employers could get there with just one click.

We also tested whether certain behavioral marketing components—message, format, and distribution method—incorporated into a brochure could increase demand for consultation services. We tested each marketing component separately. This rigorous design enabled us to test:

• the behaviorally-informed marketing as a whole compared with both no marketing and with an existing informational brochure that did not use behavioral nudges
• each message and format compared with every other message and format
• whether the email had an incremental impact on the marketing’s success, above and beyond the letter.

The study’s large sample size (97,182 worksites in 48 states and Washington D.C.) meant that we could detect small differences among different kinds of marketing.

We found that marketing via brochures was effective overall, nearly doubling the rate of requests for services compared with no targeted marketing. However, the behaviorally-informed materials performed no better than OSHA’s existing informational brochure. Thus, it appears that the brochures’ success was not due to the cleverly-designed behavioral nudges, but rather due to something common to all the brochures such as the basic informational content or the friendly cover letter from Department of Labor’s assistant secretary, which accompanied each brochure.

Why our nudges were not effective is unclear and would require further study. Would a different medium, such as TV or radio, have been more compelling? What seems certain, however, is that printed or emailed brochures are not the right medium for conveying the kinds of emotional behavioral appeals we tested in this study.