In an adaptive sampling strategy based on multi-stage sampling, information gathered in earlier stages is used to choose what information to search for in the next stage. This should be the information that has the greatest probability to best improve the basis for the decisions to be made. Based on literature this paper discusses how to use this strategy to improve investigations of contaminated soil by assessing decision uncertainty. The measurement error, the number of samples and choice of interpolation method are among the factors that have large effect on the usability of the outcome of a multi-stage sampling investigation using interpolation. In the light of increased usage of field-portable test instruments the interaction of measurement error and number of samples needs to be further investigated. Other issues that need attention are how to measure the quality of the basis for decision and how to decide when the quality is good enough.