This article proposes an approach to determine the level of Pseudomonas spp. in milk, based on the evaluation of the content of oxygen and carbon dioxide produced in the headspace of sealed vials; the research was divided into two phases: model building and preliminary validation. Three different strains of Pseudomonas spp., Ps. putida (wild strain) and Ps. fluorescens (wild and collection isolates), were used as targets. Data of CO2 and O2 were modelled through a modified positive (CO2) or a negative Gompertz equation (O2) to estimate the Minimum Detection Time (MDT), defined as the time to attain 3% of CO2 (MDT1) or a decrease in the content of O2 by 3% (MDT2). Then, MDT1 and MDT2 were submitted to a linear regression procedure, using cell concentration as independent variable; the correlations ‘MDT1/cell concentration’ and ‘MDT2/cell concentration’ showed high determination coefficients (>0.983). Moreover, the regression procedure pointed out that both MDT1 and MDT2 decreased by ca. 3 h for an increase in cell count of 1 log cfu mL−1. Preliminary validation in milk pointed out that the error associated with the regression line ‘MDT2/cell concentration’ was below 5%.