Objectives: The objectives of the study were: (i) to survey staff perceptions of causes of delay in patients' journeys through the ED; (ii) to identify and analyse key constraints to patient flow using real-time diagnostic/decision support software (Patient Flow Study); and (iii) to assess the correlation between staff perceptions and data from the Patient Flow Study.Methods: ED and non-ED staff were surveyed prior to the Patient Flow Study. The survey involved ranking the likely reasons for delay at three set points after triage (160, 320 and 480 min). Real-time data on delay in patients' journeys through the ED were collected over a period of 5 weeks. The correlation between staff perceptions and study data at the three time points was calculated using Spearman's rank correlation coefficient. Subgroup analysis was performed on the basis of staff position, years of experience at St George and whether they had previously attended training on constraints to flow.Results: A total of 68 staff responded to the questionnaire (response rate 42%). During the study period, 4555 ED attendances (97% of all presentations) were analysed for causes of constraint. Strong correlation between staff perceptions and real-time data was only found among some subgroups at the point 160 min from triage.Conclusions: Overall, staff perceptions regarding causes of constraint to patient flow do not correlate well with data obtained from real-time analysis.