Computerised neuropsychological assessments are rapidly developing on the back of technological advances such as tablet computers, smart phones and remote web-based testing. However, it is critical that the same rigorous standards in psychometric test quality are applied. The CogSCAN study aims to provide a systematic cross-comparison and validation of four prominent computerised cognitive test batteries, namely CogState, Cantab, CBS and NIH Toolbox in the older adult population including those with and without cognitive impairment. This paper describes the CogSCAN protocol including recruitment strategies, measures, methods, progress and characteristics of the first 105 community participants. Participants are assigned to a 1-month test-retest reliability arm (computer tests) or a validity arm in which they receive computer and traditional tests (at baseline and 12 months later). In total, we plan to recruit 500+ participants including community-living older adults and patients diagnosed with Mild Cognitive Impairment (MCI) and dementia. We will also identify the most sensitive and specific measures for identifying persons with MCI and dementia, investigate acceptability in this population and the influence of computer familiarity and attitudes on test performance, the culture fairness of computer tests and the sensitivity of computer tests to the earliest manifestations of cognitive decline.