AbstractWe characterise the Millimetre Astronomy Legacy Team 90 GHz Survey (MALT90) and the Mopra telescope at 90 GHz. We combine repeated position-switched observations of the source G300.968+01.145 with a map of the same source in order to estimate the pointing reliability of the position-switched observations and, by extension, the MALT90 survey; we estimate our pointing uncertainty to be 8 arcsec. We model the two strongest sources of systematic gain variability as functions of elevation and time-of-day and quantify the remaining absolute flux uncertainty. Corrections based on these two variables reduce the scatter in repeated observations from 12%–25% down to 10%–17%. We find no evidence for intrinsic source variability in G300.968+01.145. For certain applications, the corrections described herein will be integral for improving the absolute flux calibration of MALT90 maps and other observations using the Mopra telescope at 90 GHz.