Monitoring foot skin temperatures at home have been shown to be effective at preventing the occurrence of diabetic foot ulcers. In this study, the construct validity of using >2.2°C difference between contralateral areas on the foot as a warning sign of imminent ulceration is explored. Thirty participants with diabetes at high risk of ulceration (loss of protective sensation and previous ulceration and/or amputation) monitored their foot temperatures at six sites, four times a day for six days using a handheld infrared thermometer. Walking activity, time of day, and environmental temperature were also monitored and correlated with foot temperatures. We found that contralateral mean skin temperature difference was 0.78°C at baseline. At single sites, left-to-right temperature differences exceeding the threshold were found in 9.6% of measurements (n = 365), which reduced to 0.4% when individually corrected and confirmed the next day. No correlation was found between contralateral temperature differences and activity, time of day, and environmental temperature. We conclude that using a >2.2°C difference is invalid as a single measurement in people at high risk of ulceration, but the construct validity is appropriate if both individual corrections and next day confirmation are applied.