Capillary isoelectric focusing (CIEF) is an important tool for the quality assurance of biotechnologically maintained drugs and for proteome analysis. The critical performance parameters of this technique are the precisions of isoelectric point (pI) values and peak areas. Compared to capillary zone electrophoresis (CZE), where precise results can be obtained (e.g., 0.5% relative standard deviation (RSD) for peak areas, n = 60), only few data are available for CIEF experiments. So far, reproducible data of pI values (RSD = 0.5%) have been acquired, but peak areas show inferior results (about 3–15% RSD). Nonstable capillary coatings and protein adsorption have been discussed as possible reasons. Recent work of Righetti et al. [25, 27] has proven that the use of coated capillaries can reduce the adsorption of proteins by 50% but cannot prevent it. In our CIEF experiments irregular and poorly reproducible peak patterns have been observed. In a long-time experiment of 106 repeated runs, an overall RSD of 10% was obtained for peak areas, RSD of 2% only in series of about 10 consecutive replicates. Especially at higher concentrations the reproducibility deteriorates. This seems to be the result of a self-amplifying process, induced by adsorbed protein molecules, leading to further agglomerations. CZE control experiments in linear polyacrylamide (LPA)-coated capillaries proved a strong pH dependency of these effects within a small range. Compared to bare fused-silica surfaces, adsorption effects are reduced but not inhibited. An enhancement of reproducibility in CIEF experiments can be achieved only by controlling the interactions of proteins and capillary walls.