AIMS--To establish a mechanism to examine the components of turnaround time in a representative cross-section of laboratory users; and to identify potential areas for improvement. METHODS--Information was collected manually from result reports received by eight laboratory users: three wards in the main hospital, four GP practices, and one local psychiatric hospital. This was combined with data from the departmental computer files to create a spreadsheet detailing different time points in the processing of a specimen, from venepuncture to receipt of result report. RESULTS--At the main hospital, 80% of samples arrived within two hours of venesection and 95% by four hours; 75% of samples were analysed within two hours; 85% of results arrived in the wards within six hours of printing, although 12% took more than 18 hours to arrive; median overall time six hours. At the satellite (psychiatric) hospital, all samples arrived within seven hours of venesection; 45% were analysed within two hours--the rest the following morning; there were highly variable post-analytical times, minimum 18 hours, maximum 122 hours; the median overall time was 69 hours. Twenty five per cent of samples from GPs took more than 20 hours to arrive; 75% were analysed within two hours, the rest took over 18 hours--waiting overnight; the post-analytical times were highly variable, minimum 22 hours, maximum 122 hours; the median overall time was 50 hours. CONCLUSIONS--The method is easily repeatable and demonstrates the need for local improvement in the post-analytical period. Although specific to the individual data handling system for one laboratory, this method may be used as a basis for other laboratories in pathology disciplines to undertake a representative assessment of turnaround times for different groups of laboratory users.