The surface plasmon resonance (SPR) biosensor was recently introduced to the analytical biochemical society for measuring small drug-protein interactions. However, the technique has many times been used without specifying the type of enantiomeric form of the chiral drug measured and/or with using a too narrow drug concentration range resulting in biased values of binding coefficients and sometimes even assumptions about single-site bindings although the binding in reality comprises a multisite interaction. In this study we will give guidelines for reliable experimental and methodological approaches to avoid these pitfalls. For this purpose, we also introduce a new tool, based on physical chemistry, to the sensor community; the calculation of the adsorption energy distribution (AED). The AED-calculations reveal the degree of heterogeneity directly from the SPR raw data and thus guide us into a narrower selection of probable models before the rival model fitting procedure. We demonstrate how to measure reliable equilibrium data for the two typically different cases: drug binding to (i) transport (plasma) proteins and to (ii) a target protein. Both the binding of the chiral beta-blocker propranolol to alpha(1)-acid glycoprotein (AGP) and that of the anticoagulant warfarin to human serum albumin were heterogeneous, with a few strong enantioselective sites and many weak nonselective sites. We also demonstrate how the multisite binding rapidly falsely turns to single-site as the concentration range is narrowed and how adding dimethyl sulfoxide to the buffer affects multisite drug-protein data. The binding of the enantiomers of the thrombin inhibitor melagatran was investigated on both thrombin and the transport proteins, revealing clear enantioselectivity for thrombin in favor of the active enantiomer, but almost similar binding properties for both enantiomers to the transport protein AGP. The AED-calculations verified that both these system has a unimodal energy distribution and are best described with a homogeneous adsorption model.