We present the results of a web survey presented to non-respondents of a main telephone survey and compare declared mobility of both survey modes. After a description of the population who answered online, we summarise travel patterns and estimate a selection bias. Results are consistent with the hypothesis of an under declaration attributable to the web, as far as we know, that the risk of omitting trips concerns especially short trips and less constrained trip purposes. From comparison with a previous face-to-face and web survey we give some perspectives for future household travel surveys.