Many new assistive input systems developed to meet the needs of users with functional impairments fail to make it out of the research laboratory and into regular use by the intended end users. This paper examines some of the reasons for this failure and focuses particularly on whether the developers of such systems are using the correct metrics and approaches for evaluating the functional and social attributes of the input systems they are designing. This paper further focuses on the importance of benchmarking new assistive input systems against baseline measures of useful interaction rates that take allowance of factors such as input success/recognition rate, error rate, correction effort and input time. By addressing each of these measures, a more complete understanding of whether an input system is practically and functionally acceptable can be obtained and design guidance for developers is provided.