The main goal of this study was to do a heuristic evaluation of the information architecture (IA) of the recently redesigned academic library website of the University of Pretoria and to provide recommendations for improvement if necessary. The main problem that is addressed by this study is that there is no comprehensive list of heuristics for the purpose of evaluating the IA of academic library websites. This study used various methods to address this problem and achieve the main goal. A literature review resulted in a set of integrated usability heuristics that can be used to do a heuristic evaluation of the IA of academic library websites. The set was created based on an integration of existing usability principles from authorities in the field of usability. The integrated set of existing principles was made applicable to academic library websites through the addition of sub-criteria based on a review of twenty related studies. A multiple-case study method was used for the heuristic evaluation of the library websites of the top three universities in South Africa, i.e., University of Cape Town, University of the Witwatersrand, and Stellenbosch University. The results of the multiple-case study method were used to further adapt the list of principles. The heuristic evaluation method was used in each of the case studies as a data collection method. Another case study was done, using the same method and revised heuristics, to evaluate the usability of the library website of the University of Pretoria and to provide recommendations. The main outcomes of this study is a list of heuristics that can be used for the heuristic evaluation of the IA of academic library websites, as well as a list of recommendations for the improvement of the newly redesigned library website of the University of Pretoria. The list of heuristics is not limited to academic library websites, however, as it can be adapted and reused for the heuristic evaluation of other types of websites. / Dissertation (MIS)--University of Pretoria, 2017. / Information Science / MIS / Unrestricted