The aim was to evaluate subjectivity (using inter- and intraexaminer repeatability), the effect of examiner experience, and residual caries diagnostic accuracy with visual tactile (VT) criteria and using a caries disclosing agent (CD). Thirty teeth with occlusal caries were excavated by a single operator not involved in the diagnostic part of the study. A test area was marked in each cavity. Four dentists with more than five and five dentists with less than five years' experience rated the marked area twice (a week apart) using VT criteria. A week later, the samples were stained using Caries Detector. The same examiners rated the presence or absence of stain in the marked area twice (a week apart). Undecalcified thin slices were examined for bacteria using light microscopy. Overall kappa for inter-examiner repeatability was higher for CD (0.45) than VT (0.31). In the less experienced group the kappa value was higher for CD (0.41) than for VT (0.23). In the experienced group kappa was lower for CD (0.43) than for VT (0.46). Median kappa for intra-examiner repeatability was higher for caries detector (0.77, 0.53) compared to visual tactile (0.52, 0.34) for the more and less experienced examiners respectively. There was no significant difference between the experienced and the inexperienced group in combined sensitivity and specificity (mean) for VT (0.52, 0.53) or CD (0.60, 0.58). In conclusion: VT was more subjective than CD, except for experienced examiners who had a higher inter-examiner repeatability for VT than CD. Diagnostic accuracy for residual caries does not increase with experience.