Urea represents a common form of organic nitrogen (N) which is added to agricultural soils in large quantities in both cropping (e.g. fertiliser) and livestock (e.g. urine) systems. In addition, there is a small, dynamic ambient pool of urea in soil associated with metabolic functioning in the microbial community. The diacetyl monoxime (DAM) colorimetric method is routinely used to quantify urea in soil, however, it lacks specificity due to the potential to react with the ureido group (R1NHCONHR2), a common structural moiety in soil organic matter. The aim of this study was therefore to critically evaluate the accuracy of this method for urea determination in soil. Using the DAM assay, we demonstrated significant cross-reactivity with a range of ureido compounds, many of which are ubiquitous in soil. We conclude therefore that the DAM assay is highly likely to overestimate urea concentrations in environmental materials. Such overestimation was confirmed using high resolution HPLC-Orbitrap MS to quantify urea in grassland soils using standard addition and the concentrations compared with those of the DAM assay. The results obtained show the DAM colorimetric method overestimated urea concentration by between 7.2 and 58 times for the sites studied. This significant overestimation of urea emphasises the need to validate the colorimetric method with reference to the LC-MS assay to ensure the robustness of measured urea concentrations. On this basis we recommend that reporting of the results from the DAM colorimetric method as “urea” concentration be curtailed and reported as “ureido-N” to recognise the contribution of unknown and variable contributions from other compounds. Indeed, given the problems with quantitatively assessing the latter contributions we would recommend the DAM method is now avoided in surveys of urea concentrations in soil and the wider environment.