In order to help to ensure a high level of performance of GAGE geodetic networks, UNAVCO engineers spend a considerable amount of time maintaining and monitoring station quality factors. We focus on the GPS/GNSS networks operated by UNAVCO: PBO (1100 core stations), COCONet (143 stations), and TLALOCNet (33 stations). Currently engineers monitor several state-of-health parameters, looking for signs that the station is operating properly. Our goal is to standardize state-of-health evaluations, so that problematic stations can be more rapidly identified and the identification process can be partially automated. Station health parameter values will depend not only on station health, but also on instrumentation used and local environment. We compile median values for daily multipath and signal-to-noise ratios, obtained from teqc, and for post-processed station quality parameters, produced by the University of Nevada, Reno, and hosted on UNAVCO's ftp servers. Station uptime depends not only on the antenna and receiver working properly but also on telemetry and interference. We also look at time series, where station scatter or noisiness can indicate problems at a station. Furthermore, nonlinear motion in a time series may indicate a site issue. In some cases, station health problems are not caused by instrument issues but instead by external factors such as groundwater pumping, long-term drought, vegetation overgrowth, snow/ice on the antenna, etc. Such cases are investigated individually, and summaries are currently posted to a shared news page hosted by Google, the Geodetic Data Services (GDS) Technical News Community. By looking at station history and parameter history, we can retroactively test algorithms and compare them with past maintenance. While developing standards for station health parameters should flag potential problems, flagged stations will still need to be reviewed by an engineer to pinpoint and evaluate the source of problem.