The paper introduces a statistical tool called sPlot, designed to explore data samples composed of multiple event sources. sPlot is a technique that unfolds the contributions of different sources to the distribution of a given variable in a data sample. The method is applied within the context of a Likelihood fit, which determines the yields of various sources. The paper covers the basics of the Likelihood method and the validation of analysis techniques. It then presents the inPlot technique, a simpler but less accurate precursor to sPlot. The core of the paper develops the sPlot formalism, explaining its properties, implementation, and illustrations. sPlot is shown to be effective in reconstructing the distributions of control variables independently for each source of events, without prior knowledge of these distributions. The technique is demonstrated through simulated events and an application to branching ratio measurements. The paper also discusses the normalization, statistical uncertainties, and merging of sPlots, highlighting their advantages over inPlots. Finally, the sPlot technique is applied to efficiency-corrected yields, providing a tool for cross-checking analyses and extracting physics results.The paper introduces a statistical tool called sPlot, designed to explore data samples composed of multiple event sources. sPlot is a technique that unfolds the contributions of different sources to the distribution of a given variable in a data sample. The method is applied within the context of a Likelihood fit, which determines the yields of various sources. The paper covers the basics of the Likelihood method and the validation of analysis techniques. It then presents the inPlot technique, a simpler but less accurate precursor to sPlot. The core of the paper develops the sPlot formalism, explaining its properties, implementation, and illustrations. sPlot is shown to be effective in reconstructing the distributions of control variables independently for each source of events, without prior knowledge of these distributions. The technique is demonstrated through simulated events and an application to branching ratio measurements. The paper also discusses the normalization, statistical uncertainties, and merging of sPlots, highlighting their advantages over inPlots. Finally, the sPlot technique is applied to efficiency-corrected yields, providing a tool for cross-checking analyses and extracting physics results.