The plan for Ending the HIV Epidemic (EHE) in the United States aims to reduce new infections by 75% by 2025 and by 90% by 2030. For EHE to be successful, it is important to accurately measure changes in numbers of new HIV infections after 5 and 10 years (to determine whether the EHE goals have been achieved) but also over shorter time-scales (to monitor progress and intensify prevention efforts if required). In this viewpoint, we aim to demonstrate why the method used to monitor progress towards the EHE goals needs to be carefully considered. We briefly describe and discuss different methods to estimate numbers of new HIV infections, based on longitudinal cohort studies, cross-sectional incidence surveys and routine surveillance data. We particularly focus on identifying conditions under which unadjusted and adjusted estimates based on routine surveillance data can be used to estimate changes in new HIV infections.