Estimates of output gaps continue to play a key role in assessments of the stance of business cycles. This paper uses three approaches to examine the historical record of output gap measurements and their use in surveillance within the IMF. Firstly, the historical record of global output gap estimates shows a firm negative skew, in line with previous regional studies, as well as frequent historical revisions to output gap estimates. Secondly, when looking at the co-movement of output gap estimates and realized measures of slack, a positive, but limited, association is found between the two. Thirdly, text analysis techniques are deployed to assess how estimates of output gaps are used in Fund surveillance. The results reveal no strong bearing of output gap estimates on the coverage of the concept or direction of policy advice. The results suggest the need for continued caution in relying on output gaps for real-time policymaking and policy assessment.