The aim of this in vitro study was to determine the influence of deviations from the manufacturer's instructions for use of six adhesive systems on the bond strength of resin composite to enamel and dentin. Detailed questionnaires completed by 355 dentists in a survey of the use of adhesive systems in Denmark were analyzed, and the types of deviations from the manufacturer's recommended clinical procedures were categorized for the six most commonly used adhesive systems (Scotchbond Multi-Purpose, Scotchbond 1, Prime&Bond 2.1, Syntac, Gluma Classic, and Gluma CPS). Plane tooth surfaces were treated with one of the six adhesive systems either according to the manufacturer's instructions or with the deviations identified (eg, shortened application times, omission of second layers). Resin composite, which varied with the adhesive system, was then bonded to the treated surfaces. After storage in water at 37°C for 7 days, the bonded specimens were tested to failure to determine shear bond strength (n = 6). Certain deviations from the manufacturer's protocols significantly reduced bond strength to dentin for three (Scotchbond Multi-Purpose, Scotchbond 1, and Prime&Bond 2.1) of the six adhesive systems, while no negative effects were found on bond strength to enamel for any of the six systems. The adhesive systems varied regarding sensitivity to deviations from the instructions for use. Bonding to dentin was more technique-sensitive than bonding to enamel. For the most reliable result, the clinical procedure recommended by the manufacturer should be followed meticulously.