Coil overlap occurs when random coil polysaccharides such as cereal beta-glucan or galactomannan in solution are abundant enough and large enough to entangle with one another to form networks. It was recently shown that this concept applied to in vitro digested cereal-based foods could predict the efficacy of the food to reduce postprandial glycaemia. In the current study we further investigate the role of coil overlap for prediction of glycaemic and insulinaemic responses using four guar fortified breads (10-15% wheat flour replacement level) with galactomannans of different weight-average molecular weight (Mw). The breads, including a wheat flour control, were tested in a randomised crossover study in 12 overweight adults. Addition of guar reduced postprandial serum insulin, but not glucose responses. The extent of postprandial insulin reduction correlated with the solution properties of galactomannan after in vitro digestion. A significantly greater reduction in insulin response was observed for two of the breads where the galactomannan Mw and concentration in solution after in vitro digestion was above coil overlap, in contrast to two other breads, which resulted in digests containing galactomannan below coil overlap and a significantly lower reduction of postprandial insulin. Further in vitro digestion experiments focusing on amylolysis of starch with kinetic modelling showed a greater proportion of slowly digested starch in breads with galactomannan above coil overlap than below. A combination of the molecular weight of dietary fibre in a food and its soluble concentration are key parameters explaining its physiological efficiency in the upper gastrointestinal tract.