Purpose <p>Front-of-package labelling (FOPL) schemes have been developed to provide more useful information to consumers and facilitate more critical and healthier food choices. This single-blinded multi-arm parallel-group randomised controlled trial aimed at assessing the effect of different FOPL schemes on the objective understanding of the nutritional content and intention to purchase products, in El Salvador.</p> Methods <p>Participants (<i>n</i> = 1204) were exposed to two-dimensional images of 15 mock-up products presented at random and balanced orders. Participants were exposed to mock-ups featuring no FOPL scheme (control group, <i>n</i> = 305) or one of the following schemes: black octagonal warning labels (OWL) (<i>n</i> = 302), traffic-light labelling (TFL) (<i>n</i> = 297), or guideline daily amounts (GDA) (<i>n</i> = 300).</p> Results <p>The odds for choosing to purchase the least harmful of the options more often was the highest in the OWL group compared to the control group: two times higher in the OWL group (OR 2·21, 95% confidence interval 1·58 − 3·11), and 49% higher in the TFL (1·49, 95%CI 1·06 − 2·11), with no changes in the GDA (1·06, 95%CI 0·75 − 1·50). OWL also resulted in the highest odds for correctly identifying the least harmful option (OR 3·77, 95%CI 2·79 to 5·09), and for correctly identifying a product with higher amounts of sugars, sodium, total fat, and/or saturated fats (3·26, 95%CI 2·40 to 4·43).</p> Conclusion <p>OWL outperformed GDA and TFL in its ability to improve objective understanding of nutritional information and purchase intention. Results support the adoption of OWL in El Salvador.</p> Trial registration number <p>ISRCTN 12,389,597.</p> Date of registration <p>17 July 2023 (retrospectively registered).</p>