One lasting legacy of the COVID-19 pandemic may be more students attending online classes at both traditional and fully online colleges. The efficacy of online education remains unproven, especially for students with academic or other challenges. And yet, both students and policymakers lack the data they need to assess whether online classes are serving students well. Online classes are the best, and often only, option in the midst of a pandemic. Some online programs have shown promise at expanding access, helping more students learn at a lower cost. However, most online programs remain unproven, particularly on a large scale, and for students with gaps in academic preparation or other needs. Despite the risks, it is challenging to assess the quality of online education. Key sources of data on colleges do not identify whether programs are online, making it impossible for students to make informed choices or for regulators to identify potential problems. Changes in data collection are needed to protect students, including data on the graduation rates, loan defaults, and other outcomes of online students. The author recommends that the U.S. Department of Education: (1) Track colleges that move online through pandemic-related regulatory waivers to allow future oversight; (2) Collect data on online students in key student aid and higher education databases and publish key findings; and (3) Identify potentially problematic online programs and institutions by monitoring rapid growth, low student retention rates, and predatory recruiting practices in order to subject them to further scrutiny.