Abstract The effect of the non-gelling polysaccharide guar gum on thermal gelation of whey protein isolate (WPI) was investigated by compression testing, oscillatory rheology and differential scanning calorimetry (DSC). All samples were prepared in 100 mM NaCl, with the concentration of WPI held constant at 3.0 wt%. Under these conditions, DSC heating scans (Setaram microcalorimeter; sample mass ≈ 850 mg; 1 °C/min) for WPI in the absence of guar gum showed an endothermic denaturation process with maximum heat flow at ∼76 °C and an exothermic aggregation process at ∼87 °C. On incorporation of increasing amounts of guar gum (0.0–0.5 wt%) the exotherm moved to progressively lower temperature, until eventually, at 0.5 wt% guar gum, it was obscured by the more intense denaturation endotherm. This is attributed to segregative interactions (“thermodynamic incompatibility”) with guar gum promoting aggregation of thermally denatured whey protein. After heating to 80 °C and cooling to induce gelation of the whey protein, Young's modulus ( E) at 20 °C rose to a maximum at ∼0.05 wt% added guar gum and decreased at higher concentrations, until at 0.5 wt% guar gum the mixture was liquid-like. A similar initial increase and subsequent decrease with increasing concentration of guar gum was observed by oscillatory measurements (1 rad s −1; 0.5% strain) of storage modulus ( G′), but the maximum occurred at ∼0.1 wt% guar gum. The difference in optimum concentration is ascribed to use of a different time–temperature regime in preparing the samples for compression testing. The reduction in gel strength observed by both techniques at high concentrations of guar gum is attributed to excessive aggregation of WPI in response to segregative interactions, with consequent collapse of the crosslinked network. However, at the very low optimum concentration of ∼0.1 wt% guar gum a large (∼12-fold) enhancement in gel strength ( G′) was evident in comparison to WPI alone.