The term ‘anti-irritant’ (AI) was coined in 1965 by Goldemberg to describe a diverse group of topical product ingredients, which were able to reduce the irritation potential of other more irritating ingredients in the same product. ‘AIs’ are being added to cosmetic formulations in order, allegedly, to benefit tolerability of the products and allow claims such as ‘soothing’ and ‘healing’ ingredients. Limited documentation in favour of the efficacy of AIs is published. We studied the dose-related effect of 4 alleged AIs (nifedipine, (–)-α-bisabolol, canola oil and glycerol) on experimentally induced acute irritation in healthy volunteers. Each AI was used in 3 concentrations. Acute irritation was induced by occlusive tests with 1% sodium lauryl sulfate and 20% nonanoic acid in N-propanol. The irritant reactions were treated twice daily with AI-containing formulations from the time of removal of the patches. Evaluation of skin irritation and efficacy of treatments were performed daily for 4 days using clinical scoring, evaporimetry (transepidermal water loss), hydration measurement and colourimetry. Only glycerol showed dose–response and effects potentially better than no treatment. There was no significant effect and no difference between the three other AIs.