A formic acid biosensor for air monitoring has been evaluated using chemometric methods. Using experimental design eleven factors that could influence the performance of the biosensor were examined. The response matrices consisted of six parameters (steady state currents at three different formic acid concentrations and response rates during changes in formic acid concentrations) describing the performance of the biosensor. The data were evaluated using a combination of principal component analysis (PCA) and multiple linear regression (MLR). To confirm the conclusions from the PCA-MLR partial least squares (PLS) was also used. The most important factor for the biosensor performance was found to be the enzyme concentration. Using the information from the chemometric analyses the optimum operation conditions for the biosensor were determined. The steady state currents were increased by 18–30% and the initial two response rates increased by 47–89% compared with a biosensor that had not been optimised.