Nonprofits are under increased accountability pressures to demonstrate their effectiveness. Output measurement (how much is produced) is disregarded as simplistic. Emphasis is made instead on measuring outcomes (changes in the lives of beneficiaries) or impacts (effects developed relative to the mission of the nonprofit, or the overall public good), and a growing portion of organizations state that they measure these effects. However, we question the assumption that outputs such as the number of beneficiaries served are being adequately measured. We first review existing research gaps on results measurement practices and discuss the main types of obstacles to the quality and utility of evaluation data. In this context, we argue for the need to reground nonprofit evaluation in the profound knowledge available about beneficiary populations. We discuss the potential and limitations of reach, a basic output indicator that is defined as the number of individuals directly affected by a nonprofit, and explore the organizational drivers of reach measurement. Evidence from 2,229 nonprofits shows they still lack adequate data on the beneficiaries they serve, face relevant conceptual and practical hurdles when trying to identify them, and are significantly influenced by organizational factors in their capacity to track them. Our research not only shows that nonprofits fail to adequately measure outputs, but also that measuring the number of beneficiaries served and how they are served is not as straightforward as outcome and impact advocates suggest. Practitioners and funders are reminded of the need to place beneficiaries at the core of their evaluation efforts.