The current meta-analysis aimed to quantify the effectiveness of hypnosis for reducing pain and identify factors that influence efficacy. Six major databases were systematically searched for trials comparing hypnotic inductions with no-intervention control conditions on pain ratings, threshold and tolerance using experimentally-evoked pain models in healthy participants. Eighty-five eligible studies (primarily crossover trials) were identified, consisting of 3632 participants (hypnosis n = 2892, control n = 2646). Random effects meta-analysis found analgesic effects of hypnosis for all pain outcomes (g = 0.54-0.76, p's<.001). Efficacy was strongly influenced by hypnotic suggestibility and use of direct analgesic suggestion. Specifically, optimal pain relief was obtained for hypnosis with direct analgesic suggestion administered to high and medium suggestibles, who respectively demonstrated 42% (p < .001) and 29% (p < .001) clinically meaningful reductions in pain. Minimal benefits were found for low suggestibles. These findings suggest that hypnotic intervention can deliver meaningful pain relief for most people and therefore may be an effective and safe alternative to pharmaceutical intervention. High quality clinical data is, however, needed to establish generalisability in chronic pain populations.