The last decade transformed faceted navigation from a “nice-to-have” into a “must-have functionality” for all online web services that contain a search function. The movement has been started by commercial websites, such as online clothing stores, with the purpose of facilitating access to the products. Now, faceted search seems to be more and more well-known within the community of information institutions and their digital services, too, as for example the online catalogue of the Library of Congress or Europeana show. In this context, as well, the feature is often sold as an added value to access the content of their sites in a more effective way. Still, experience shows that the use of faceted navigation in a digital library context is more complex than for online clothing stores. For this reason, it is important to focus on the usability of such functionalities and to test it as often as possible during implementation and regular use, as well. Site providers have to be aware of one important fact: a faceted search that has not been implemented correctly or does not work in the way the user expects it, will create frustration and/or not be used at all. This implies in a worst case scenario that users don’t come back to the site anymore, and may even spread their negative experience among the community. To prevent this, usability tests, either with experts or real users, should be conducted regularly. This can be done with common usability tests, using direct human interaction with a system, or with online tests, where test user are free to fill in the test whenever they want. In both cases, test results illuminate a system’s weaknesses and expose dysfunctional issues, which should be improved in order to offer a positive experience to its users. For more than five years, the project ACCEPT, a subproject of a Swiss national project called e-lib.ch, analyzes the usability and usefulness of digital libraries, by using user oriented methods. Experience has shown that filters provided through faceted navigation are considered as positive and very useful by end users. Nevertheless, based on different test results, several returning mistakes have been detected and it turns out that there are some ‘unwritten standards’ concerning e.g. position, labelling or ranking which should be respected to fulfil the aim of a good usability which users do expect of such web services. In this poster we will first give an introduction to faceted navigation, actual design issues and their use in digital libraries and then present testing methods, which can be easily applied in a digital library context. Together with a list of best practices concerning faceted navigation drawn out of different test experiences, the paper should give the reader all important information to evaluate its current faceted navigation and see where improvements could be made.