Aims: To validate the effectiveness of a miniaturized most probable number method (mMPN) in enumerating Salmonella from poultry matrices. Methods and Results: A MPN was developed, based on the ISO 6579:2002 method using modified semi-solid Rappaport-Vassiliadis media as the sole selective medium. The validation of the mMPN was shown to not differ significantly from, at the 95% confidence level (Student's t-test P = 0 center dot 357) to, the traditional 9-tube MPN (tMPN) using pure cultures of Salmonella ser. Typhimurium, Infantis, Montevideo, Muenster and Salmonella subsp II 1,4,12,27:b:[e,n,x] (Sofia). The validation of naturally and artificially contaminated poultry matrices (carcasses, scald tank water, faeces, caeca and feed) showed that detection using the mMPN compared well to the ISO 6572:2002; sensitivity (92%), specificity (97%) and agreement (KAPPA 0 center dot 72). The quantitative comparison between the tMPN and mMPN methods showed that 92% of enumerations were less than +/- 1 log different (Student's t-test = 0 center dot 13). Financial analysis showed that the mMPN required 64% less media and 56% less labour than the tMPN. Conclusion: The mMPN is a consistent, easy to automate method for the enumeration of Salmonella from different poultry matrices. Significance and Impact of Study: The miniaturized MPN reduces the material and labour cost of the method and enables the uniform and accurate measurement of the effectiveness of intervention strategies in the control of Salmonella colonization of poultry.