Background: Concerns about the state of the teacher workforce are evergreen. Both teacher satisfaction and interest in entering the profession have declined in recent decades (Kraft & Lyon, 2024), while, at the same time, significant concerns about turnover and shortages have grown. There are multiple potential explanations, but a significant concern is that teachers often feel isolated and have to juggle too many tasks. The concern is not new; Lortie (1975) criticized the "egg-crate" model of schools with teachers isolated in classrooms. In response to these concerns, some systems are working to redesign the teaching job to make it more attractive and sustainable via "strategic school staffing" (Education First, 2023). We study one specific strategic school staffing model and estimate its effects on student and educator outcomes. Intervention: The intervention we study is the Next Education Workforce (NEW) initiative, a strategic school staffing model that has been growing steadily over the past five years and is now implemented in dozens of schools and districts across the country. NEW team-based teaching models are intended to break the "egg crate" model of schooling: instead of isolated classrooms, teachers share a roster of students with a team of educators with complementary skills and expertise. The NEW theory of action (Figure 1) suggests that teaming allows for differentiation and more student-centered learning environments, which in turn leads to improvements in both educator and student outcomes. Research Questions: We ask three related research questions on the impact of NEW and how the impact varies across schools and students: (1) Does implementing a NEW model improve student outcomes, including persistence, achievement, and attendance?; (2) Does implementing a NEW model improve educator outcomes, including retention and evaluation ratings?; and (3) How do these impacts vary across schools and students? There is no existing causal research on NEW, but NEW is thought to have positive impacts on educator and student outcomes based on previous empirical evidence on the positive effects of teacher collaboration (Dee et al., 2006; Ronfeldt et al., 2015) as well as descriptive work on NEW showing positive correlations between teaming and educator outcomes such as satisfaction and effectiveness (Laski, 2024). The theory of action also suggests a path from educator outcomes to student outcomes -- for example, increased educator retention and satisfaction could then lead to improved student outcomes. This would suggest we may be able to measure impacts on educator outcomes before seeing impacts on student outcomes. Setting & Population: We use data on implementation of NEW in Mesa Public Schools (MPS). MPS is the largest public school district in Arizona, with 82 schools and roughly 64,000 students. Roughly 42% of MPS students identify as Hispanic/Latino, 14% are English language learners, and 20% receive special education services. MPS has been an implementing partner with NEW since 2020, and NEW has steadily expanded across schools and grades in MPS every year since 2020. Using the most inclusive definition of teaming, 15 schools and 195 educators were implementing NEW in 2020-21, and the number has now grown steadily to 45 schools and 620 educators in 2023-24. We use outcomes for all implementing grades and compare them to non-implementing grades over time. Data: We use administrative data on students and teachers from MPS from 2017 to 2024. Our data allows us to link students to teachers by grade, subject, and year. Our key outcome measures are student test scores, student attendance, student persistence, teacher retention, and teacher evaluation ratings. Research Design: We leverage the fact that NEW has been rolled out across schools and grades in Mesa over time. This staggered implementation allows us to run event study regressions with a student-level outcome as the dependent variable and a series of binary variables for each year relative to initial program implementation in the school-grade cell. Models include year and school-grade fixed effects as well as a vector of student-level controls, including baseline achievement, special education status, English learner status, race/ethnicity, and sex. We estimate similar models for educator-level outcomes and include educator controls such as role, race/ethnicity, sex, and years of experience. For these event study designations, the comparison group is all untreated (non-teamed) school-grades in Mesa in a given year. A potential concern is that teamed grades, particularly early implementers, are systematically different from non-implementing grades. This design will account for these differences by comparing trends in outcomes before NEW was first introduced. As a robustness check, we will test to confirm that pre-treatment trends are similar (parallel) between treated and non-treated grades. We will also run specifications where we only use fully untreated grades as comparisons, removing any situations that compare early adopters to later adopters. Preliminary Findings: Our current findings include basic descriptive characteristics of NEW teams in Mesa. Table 1 presents descriptives of the makeup of teams in 2023-24, using the most inclusive definition of teams. Teams are present at all school levels and have three to four unique individuals on average. Depending on the school level, about a quarter to a third of teams include a designated lead teacher, and 4-17% of teams include a teacher candidate. Figures 2 and 3 present basic descriptives comparing teachers working in teams (and the students assigned to those teachers) to their non-teamed peers. We find that, on average, teamed teachers are much more likely to be earlier in their careers. Students assigned to teamed educators have lower achievement levels, on average, and are just as likely to identify as needed special education services as their non-teamed peers.