This paper focuses on the effect of different types of financial aid on the longer-term persistence of four-year college students who graduated from a large urban school district. The paper first presents impact findings, based on two surveys and data from the National Student Clearinghouse, of an evaluation of an intervention designed to improve the college outcomes of high school students in the Los Angeles Unified School District (LAUSD). The program was operated by EdBoost, a Los Angeles nonprofit organization, and matched high school students to paid college mentors who themselves were students in Los Angeles area undergraduate and graduate programs.  The intervention was designed to offer an alternative to the limited opportunities available to LAUSD high school students to obtain college advice from high school counselors in their schools (who have caseloads of as many as 1000 students). The study included high school students whose grade point average in their junior year was sufficiently high to where they would be eligible for enrollment in the California State University (CSU) system. The study recruited 2500 of these students throughout Los Angeles. After random assignment, 1000 of these students were offered a mentor who would help them navigate the college application process for a year. The mentors were not supposed to provide academic services but were tasked with providing the high school students with advice, support, reminders, and the kind of continuous attention that low-income students often lack during their academic careers in high school.  The program began during the spring semester of the students’ junior year in high school and concluded a year later when high school students were about to graduate and would have been in the process of selecting a college from among the ones that admitted them. Our research found a relatively modest treatment contrast between the treatment and control group but also found that students in the treatment group generally received consistent services throughout the program year. Despite this modest treatment contrast, we found statistically significant impacts on enrollment in four-year colleges, enrollment in the University of California or a CSU campus, and greater usage of scholarships and grants by treatment group students once admitted to college. We also found persistent and statistically significant effects on long-term college enrollment. The paper focuses on the role of financial aid in achieving the longer-term impacts on college persistence. It examines whether the program’s impact on access to and use of grants and scholarships mediated its impact on long-term college persistence. It explores these findings in the context of the existing literature on the impact of student financial aid on college success. It also explores whether these impacts vary along several important subgroup dimensions, including whether the students are the first in their family to attend college.