Lumpfish (Cyclopterus lumpus) are used as a non-pharmacological method to control salmon lice (Lepeophtheirus salmonis) in Atlantic salmon (Salmo salar) aquaculture. However, their effect as cleaner fish varies substantially. This study examines how alternative prey availability, individual characteristics and deployment strategy influence salmon lice consumption in lumpfish, aiming to improve decision-making and effectiveness of lumpfish-based lice control in salmon aquaculture. Stomach-content data from 28,677 lumpfish sampled across multiple sites and years in the Faroe Islands (2015-2025) were analysed. The dataset included fish size and body condition, presence of alternative prey in stomachs, and cage-level salmon lice abundance on cohabiting salmon. Most lumpfish stomachs (>96%) contained no salmon lice, indicating low treatment engagement under typical farming conditions. Due to the high proportion of zeros and large variation in salmon lice counts (0–290), a probability-based modelling approach was used to describe both the likelihood and the intensity of salmon lice consumption. The probability of salmon lice consumption increased with salmon louse abundance on cohabiting salmon and was strongly reduced when jellyfish were present in stomachs. Consumption intensity showed non-linear relationships with salmon louse abundance, lumpfish size and body condition, peaking at small sizes and optimal condition. These results indicate that salmon lice consumption is opportunistic and strongly influenced by prey availability and deployment conditions. Avoiding deployment during periods dominated by jellyfish blooms, accounting for salmon lice levels at deployment and maintaining good body condition through appropriate feeding and husbandry may improve the effectiveness and targeted use of lumpfish as a biological control agent.