Website evaluation methods are an important tool to gather information for the development and the management of a website in order to ensure a good acceptance by the users. Mainly based on detailed questionnaires administered to actual or potential users, this activity can be costly and time consuming. This paper present methods, based on the statistical bootstrapping techniques, to derive confidence intervals for the evaluations produced by a certain sample of users. It is shown that relatively small (less than 10%) confidence intervals can be achieved even with small sample sizes (less than 20).