The detailed matrices and their behaviors during pesticide residue analyses were clarified using a metabolomics analysis approach. The matrix profile was investigated using two different extraction solvents, acetone and acetonitrile. Acetone extracted the matrix components with a wide range of log P(O/W) values. Components with log P(O/W) values >10, such as sterols and tocopherols, and components with log P(O/W) values <3.2 were more extracted by acetone than by acetonitrile. In contrast, components with log P(O/W) values in the range from 3.2 to 10 were extracted by both acetone and acetonitrile at the same concentration level. The study also examined the difference in the column cleanup efficiency using a solid phase extraction (SPE). Florisil, silica gel, NH(2), PSA, and GCB were selected as representative columns for pesticide residue analysis, and acetone extraction of brown rice was selected in this experiment. Most of the matrix components were removed by either column, whereas monoacylglycerols, which are the components causing the matrix effect, were not removed by any column. Understanding such a detailed matrix behavior helps to develop a better analytical method for pesticide analysis using GC-MS.