For over 15 years our first-year engineering design program has focused on a user-centered approach to design thinking and communication, where students work with real-world clients on ill-defined problems and communicate their ideas in a variety of ways to multiple audiences. Over this time frame ~5,000 students have passed through the two course sequence, and addressed over 1,500 design challenges. Since students work in teams of four to address these challenges (and will be expected to work on project teams throughout the undergraduate engineering curriculum and later in industry), we are strongly committed to helping them develop greater competency in teamwork, as opposed to simply participating in an unguided team experience. To facilitate teamwork learning, we historically used two instruments: (1) an intra-quarter peer review and self-review and (2) an end-of-the-quarter reflective memo (benefits and limitations of this approach have been described elsewhere 1,2,3). In the fall of 2011, our first-year program partnered with the university's Center on Leadership to offer students more opportunities for teamwork reflection, peer- and self-assessment and teamwork improvement throughout the two courses that comprise the program. Students used a combination of online exercises and team meetings to create a team charter, reflect on personal and team performance, provide specific feedback to team members, and use that feedback to create goals for improving their own teamwork performance--all by the middle of each course. At the end of each course, students used peer-assessment and reflective memos to determine whether they had been successful in achieving their mid-term goals. Since all assessment would be out of class, the additional workload for the design faculty was to be minimal. The students' activities would serve as a foundational experience that could be revisited by the students and the Center on Leadership in future courses utilizing teamwork. However, at the end of the year (spring 2012), when we surveyed ~425 students in the program (162 responded), we were disappointed to learn that, while some of the students found the leadership center's activities highly beneficial, an overwhelming number saw them simply as "busy work." In addition, a majority of the program's faculty, who had originally thought that the online reflective exercises would benefit the students while reducing their workload, were also frustrated by the new tools. Although we streamlined the process for the next academic year, survey results in spring 2013 were equally disappointing. Analysis of the survey responses and the online tools activities suggested that the problem was one of balance: since teamwork is a goal of the program, but not its primary goal, there were apparently too many exercises related to teamwork, ironically undermining their usefulness. In addition, by outsourcing the responsibility for administering the activities, faculty were less involved in teamwork pedagogy, unintentionally suggesting that teamwork was not integrally related to excellence in design. We did however learn a great deal about what students see as the main causes of team failure, what teamwork skills they most want to develop, and what students mean when they talk about teamwork habits, such as delegating tasks or improving communication. After first describing our several approaches to improving teamwork pedagogy, this study reports on lessons learned and modifications we have made to move forward. Briefly, we have streamlined the number of required teamwork activities, more carefully connected them to the project work, and brought more of the activities "in-house," making design faculty more responsible for the first and last activities. Our plan is to continue assessing these areas at the end of the 2013-2014 academic year. [ABSTRACT FROM AUTHOR]