Purpose: The purpose of this paper is to empirically test a new disabled service user-specific service quality model ARCHSECRET against a modified SERVQUAL model in the context of disabled students within higher education. Design/methodology/approach: The application of SERVQUAL in the voluntary sector had raised serious issues on its portability into this sector in general and its ability to measure the experience of the disabled service user in particular. In consequence, a disabled service user-specific service quality model--ARCHSECRET--was developed which led to this research being designed to compare ARCHSECRET and a modified SERVQUAL model in terms of their ability to predict and explain the variation in the service quality experience of disabled students in higher education. Findings: ARCHSECRET was superior to the modified SERVQUAL in terms of its overall predictive power; ARCHSECRET key drivers were different and better in predictive power than those of the modified SERVQUAL; and ARCHSECRET was found to be reliable and valid for the measurement of the disabled student experience in higher education, while acting as a diagnostic tool for the identification of service quality shortfalls. Research limitations/implications: The reported research should be regarded as a pilot study whose results are worthy of further investigation among larger samples of disabled service users. Originality/value: It is held that the disabled service user-specific ARCHSECRET model has made a positive contribution to the measurement of service quality within the context of disabled students in higher education while demonstrating its superiority over the SERVQUAL scale which did not quite "measure up". (Contains 11 tables and 3 notes.)