Objective: To determine the influence of tutor expertise on the uptake of a physiotherapists' educational programme intended to promote the use of outcome measures in the management of patients with stroke. Design: Pilot randomized controlled trial. Methods: Thirty physiotherapists involved in stroke management were randomized into two groups and participated in five tutor-guided educational sessions (the Physiotherapists' Educational Programme on Clinimetrics in Stroke, PEPCiS). Groups differed from each other with respect to tutors: one experienced and one inexperienced in stroke care. Primary outcome was 'actual use' (the frequencies of data of seven recommended outcome measures in the patient records of the participating physiotherapists). Results: The actual use of instruments shifted from a median of 3 to 6 in the expert tutor group and from 3 to 4 in the non-expert tutor group (P = 0.07). Physiotherapists educated by the expert tutor used a broader variety of instruments and appreciated the educational programme, their own knowledge gain and all three scales of tutor style aspects significantly more than their colleagues of the non-expert tutor group (all P<0.05). Univariate analysis on the entire set of data revealed eight factors, including tutors' performance, that were associated with a change score of the use of two or more outcome measures by individual physiotherapists after the educational programme. Conclusion: In this pilot trial it was not proven that tutor expertise in stroke care influences the actual use of outcome measures, but it warrants a future study with sufficient power to investigate the influence of the tutor. © The Author(s), 2009.