The feeding ecology of wildlife populations has important implications for individual health, population productivity and distribution patterns. For ursids (bears), food resources and feeding behaviour primarily affect population dynamics via effects on cub production and survival. Much of what is known about the feeding ecology of bears is based on analyses of tissues collected from capture-based research efforts, harvested animals or non-invasive approaches. However, inference about diet from hair has been limited by a lack of quantitative data on the timing of the moult and hair growth rates. We conducted a study to develop and test two methods of quantifying hair growth rates of three species in the family Ursidae (n = 1 polar bear, Ursus maritimus; n = 3 black bears, Ursus americanus; n = 3 grizzly bears, Ursus arctos horribilis). We implemented visual and biochemical approaches, proven safe for humans and other mammals, in a zoo setting. These methods relied on voluntary bear behaviours trained using positive reinforcement. The two methods were: (i) applying a small patch of hair dye (or bleach) on the rump or foreleg, and (ii) feeding an isotopically labelled amino acid (glycine) capsule that ‘marks’ time at a particular location as it is incorporated within the hair. We collected hair at regular intervals (every 1–2 weeks) for five months from body locations on the bear consistent with commonly sampled collection points in wild-caught bears. We found that both methods effectively identified periods of hair growth and detected individual and seasonal variation in hair growth rates. Average guard hair growth rates ranged between 0.10 and 1.05 mm day⁻¹ across the three species. This study provides the first step for developing a foundation for incorporating seasonality in wild-collected bear hair samples by assessing growth over an annual cycle.