Wildlife tracking studies have become ubiquitous in ecology and now provide previously unobtainable data regarding individual movement, vital rates, and population demographics. However, tracking devices can potentially reduce survival of study subjects, generating biases in the vital rates they seek to measure. Previous studies have found that greater sage-grouse (Centrocercus urophasianus) fitted with Global Positioning System (GPS) tracking devices may experience reduced survival, relative to those tracked with traditional radio transmitters, and have documented skin abrasions and lacerations associated with typical backpack-style GPS harnesses. We implemented an experimental study comparing survival and harness retention between 2 different backpack-style GPS transmitter harnesses. We captured female sage-grouse at 3 study sites in the northwest Great Basin of Oregon, Nevada, and California during 2019–2021. We fit each individual, following previously published recommendations, with either a standard backpack harness or a modified harness hypothesized to reduce skin abrasion and laceration. We used known-fate models in Program MARK to model variation in survival and harness retention separately as a function of harness type, year, age, a linear effect of time, and the ratio of the device to individual body mass. Neither survival nor retention varied systematically by harness type, however retention decreased as a function of body mass ratio. We echo previous recommendations for standardized harness attachment protocols and studies designed to isolate and test potential mechanisms by which tracking devices and attachment methods might affect survival and well-being of sage-grouse and other tracked species.