A semi-empirical model is developed to predict the maximum achievable strength of a nonwoven fabric and to identify and scale the factors affecting that strength. The scope and accuracy of the model are limited by the assumptions made during its development, but they are shown to be quite reasonable. The results of the model calculations suggest that the practical potential strength of an ideal point-bonded nonwoven fabric is three to four times higher than the average tensile strength of the best available spunbonded fabric samples, and about one-third lower than the average strength of fiber samples. Weak fiber-bond interfaces, areal density variations, fiber property variations, and fiber curl or combinations of these and others factors may limit fabric strength. Of these four factors, the first two are the key ones that limit spunbonded fabric observed strength. Knowledge of the potential strength and the factors limiting strength may enable manufacturers to improve nonwoven strength.