What happens when organizations start measuring their knowledge workers like gig workers on an Uber platform or production workers in an Amazon warehouse? This research describes organizations using algorithmic control mechanisms to evaluate their knowledge workers. Based on 68 interviews, our research identifies how algorithmic control impacts knowledge workers. We identified the following impacts: (1) standardization promotes fairness but creates compliance costs, (2) not measuring the full scope of work, and (3) metrics focused on short-term productivity over long-term growth. Our research extends prior work on algorithmic control of gig and warehouse workers to knowledge workers in professional jobs. Contrasting prior work, our research advocates for a middle space where algorithmic control can coexist with human discretion and innovation. In this middle space managers assume boundary-spanning where they have the discretion to override algorithms in support of employee innovation and can help employees make sense of the algorithms.