A literature review of sampling strategies for contaminated soil and groundwater is presented. The different types of uncertainties associated with a site-investigation are reviewed. The un-certainties are classified as pre-sampling, sampling, and post-sampling uncertainties. Theories to quantify different types of uncertainties are reviewed, especially the different sampling uncertainties. The particulate sampling theory was found to be of interest because it identifies all the relevant sampling uncertainties and presents them in a structured way. The question on how many samples to collect under different conditions is addressed. A conclusion is that the sampling objective must be clearly stated prior to the development of any sampling plan. Two of the most common sampling objectives are to detect hot spots and to estimate mean concentrations. Methods to determine the economic worth of sampling are reviewed. Traditionally, sampling has been performed with the objective to either (1) minimize uncertainty for a fixed sampling budget, or to (2) reach a prespecified accuracy at lowest possible cost. However, in recent years data worth analysis has been used increasingly. The valuable feature of data worth analysis is the ability to assess the worth of a proposed site-investigation program prior to performing any sampling or measurements. Data worth analysis is performed in a risk-based decision framework by combining decision theory with e.g. geological, geochemical, hydrogeological, and economic information. A conclusion is that the use of data worth analysis for contaminated land problems has a potential but the complexity of existing methods has limited its use. To be used on a more regular basis, data worth methods need to be further developed. A review of more than 30 related software packages is presented in an appendix. It was found that public domain or low cost commercial software packages exist for a number of applications related to site-investigations, e.g. sampling design, data evaluation, geostatistical techniques, stochastic simulation, data worth analysis, and decision analysis.