Paid TV schedules were placed, and additional \"value added\" public service announcement (PSA) schedules were negotiated for high-ozone-level months. The commercials were also formatted for sharing and promoting through Facebook's video tab. Each month, a percentage of the budget was allocated to expand the reach of the commercials to the Facebook audience. Over the course of the campaign, the project team managed a variety of key metrics and quantiable goals. These included: Digital Key Performance Indicators (KPI's) - Increase in web trafc and increase in social media. Digital Posting Target Ratio: For every 10 social posts, there was a content ratio of 6:3:1. Six posts focused on environmental news to help establish the Little Steps. Big Impact. page as a source of thought leadership. Three posts focused on engagement by asking questions, providing shareable content, or encouraging feedback. One post directly solicited followers. Re-post Testing: During cooler months when non-attainment risk was decreased, previously approved posts were compiled, edited for timeliness, and rescheduled to post two to three times a day to test impact of an increased posting frequency. Results of this testing showed that frequency increased message reach (number of people exposed to the message), but without a corresponding increase in engagement. paid placement that could target specic audiences across more than 200 variables in demographics, interests, and behaviors. By logically choosing those variables, the team hoped to nd those individuals most likely to be receptive to messages and willing to engage with the effort. The project team first gee-targeted the two counties comprising most of the metro area, then selected for interests including health, wellness, environment, air pollution, environmental protection, air quality, commuters, park and rides, bike trails, and healthy living. Initial paid posts were AXB tested, which is sometimes referred to as split testing and allows the comparison of results between two versions of an ad to see which ad or variables result in better performance. Tests included running the same message with different images. For example, in posts promoting carpooling, one featured an image from the campaign's television commercial, while a second featured a stock image of heavy trafc. Through AIB testing, the team learned that campaign images typically outperformed stock images, on average, 3:1 for engagement actions. UNHEALTIW n SENSWIVE GROU PS To help raise awareness of the day-to-day changes in air quality, a continuum of emoji 's were developed that graphically represented the level of harmJl ground-level ozone measured at a central location. The emoji's were updated daily as air quality changed, and shared through both social media and on electronic outdoor boards throughout the metro area. Television commercials supported the emoji messaging while also giving examples of \"little steps\" that could be taken to improve the air quality and contribute to pushing the emoji measurement toward the positive. Guide to Air Quality Index Categories UNHEAI.TH\\Ir FOR UNHEALTHY VERY HAZARDOUS SENSITIVE GROUPS \"HEALTHY