Spatially nonlocal regularization techniques invariably introduce one or more material length parameters. A major difficulty resides in the experimental determination of such intrinsic length scales. Several indirect techniques have been proposed in the past years which are based on the observation of macroscopic mechanical effects which ensue from the process zone, e.g. the size effect. The present contribution uses a different approach by exploiting local measurements which were carried out in and around the process zone. Using an appropriate experimental technique, displacement and strain fields have been determined during the damage and fracture process in a CT-test of a composite material. The unknown length scale is extracted by comparison of the experimental results with the numerical simulations which have been obtained with a transient gradient damage model. The necessity and the importance of such a local comparison is motivated and discussed. A parameter estimation based on the global mechanical response of one experiment only, may lead to an erroneous and deluding determination of the intrinsic length scale.