Museums, galleries and other cultural organisations have been swift in their adoption of the Web and virtual visits to museum Web sites have become popular. For many smaller museums the cost of developing and maintaining a Web site is difficult to justify, particularly where the economic benefit to the museum is hard to demonstrate. Many sites are therefore developed as unofficial, in-house projects—often without financial support. Evaluation is essential for determining whether a Web site is meeting the needs of its users and should be part of an ongoing process, from initial conception to long-term maintenance and development. This contribution focuses on the evaluation of Web sites as a way of addressing some of the limitations inherent in the non-professional development environment, placing it within the context of current museum Web development practice. The paper presents a case study illustrating Web evaluation issues from a computing perspective, using methods appropriate to the non-professional development environment. Four evaluation methods are examined in detail: direct observation, log analysis, online questionnaires, and inspection methods. The experience of applying each method, the benefits and limitations of each method and the relative effectiveness of the methods are analysed.