Most assessment arguments are about measurement. When is it better to use direct versus indirect measures of student learning? Is one standardized test of critical thinking better than another? Is applying rubrics to student work better than using standardized tests? How valid are self-reported measures of learning? Although these arguments are important, like most academic arguments they focus on theoretical and methodological purity rather than the more practical (and, in the authors' view, more important question): How can we work with assessment evidence to create changes that improve student learning? Thoughtful research and discussion on the validity of different kinds of assessments are important, but the authors believe that the heavy focus on these issues leads one to assume that the biggest challenge to using assessment to improve student learning is technical, and that once one creates sufficiently good measures, widespread institutional improvement in student learning will follow. To the contrary, the authors believe that most assessment methodologies already offer "good enough" evidence to promote changes that can benefit students. The real challenge is to find ways of using admittedly imperfect evidence to create improvements within the complex governance, social, political, and value structures that mark campuses. This chapter details the lessons the authors have learned and the good practices they have either created or appropriated in their efforts to help institutions participating in the Wabash National Study of Liberal Arts Education (the Wabash Study) use evidence from the study to improve student learning.