Objectives An analysis of the trough serum concentrations sent to the UK Antimicrobial Reference Laboratory for teicoplanin therapeutic drug monitoring (TDM). Methods All trough concentrations over a 13 year period were analysed and the percentages were calculated for the following: <10 mg/L (a sub-optimal concentration for all); ≥10-<20 mg/L (the target used for ordinary Gram-positive infections); ≥20-<60 mg/L (the target for all severe staphylococcal infections including endocarditis); and ≥60 mg/L (the concentration associated with toxicity). Results The percentage of patients with concentrations of <10 mg/L decreased each year to 13% in 2006. Almost 40% of the samples each year were in the ≥10-<20 mg/L range. In 1996, the percentage of samples in the ≥20-<60 mg/L range reached a study high of ∼70%. That percentage then fell to 30% and increased slowly to 50% at the end of the study. Fewer than 5% of the samples were ≥60 mg/L. Conclusions Our study shows that there is a need to increase the initial dose or extend the number of days that the loading dose is used in a significant number of patients. With such a wide optimal range and a low potential for toxicity, it is unclear why optimal therapy is not achieved in a higher percentage of patients.