Abstract When introduced into a new herd, goats are confronted with unfamiliar animals. Their behavioural and physiological reactions during this confrontation are likely to differ depending on the presence or absence of familiar conspecifics (peers). To assess these reactions, we confronted 12 goats both alone and with two peers (confrontees) with established groups (n = 4 groups) consisting of goats unfamiliar to the confrontee (unfamiliar goats) (12 goats × 2 confrontations = 24 confrontations in total). Each confrontation lasted for one hour. Agonistic interactions, sniffing behaviour and level of activity were recorded throughout the confrontations. In addition, concentrations of cortisol metabolites were measured in faecal samples taken in the evening before the confrontation and three successive samples after the confrontation. Before the start of the experiment, we evaluated the dominance relationships of the involved goats within their respective housing groups by direct observations made during the main feeding times. Data were analysed using generalised linear mixed-effects models with the fixed effects presence of peers (yes, no), rank category (high, medium, low) and repeated confrontation (numeric variable). For the analysis of activity level and concentrations of faecal cortisol metabolites, period (minutes 0–15, 16–30, 31–45, 46–60) and sample (control, 13, 14, 15 h after the confrontation), respectively, were included as additional fixed effects. Unfamiliar goats directed fewer agonistic interactions towards confrontees when the latter were accompanied by peers compared to when they were alone (without peers: 57; with peers: 20 interactions per animal and confrontation). The same was true for the proportion of agonistic interactions involving physical contact (without peers: 69; with peers: 53%) and the number of sniffing behaviours (without peers: 16; with peers: 9 interactions per animal and confrontation). On the other hand, confrontees with peers were more likely to direct agonistic and sniffing behaviour towards unfamiliar goats than those on their own. Confrontees with peers had lower concentrations of faecal cortisol metabolites after confrontations (without peers: 273; with peers: 198 ng/g). For confrontees (with and without peers), activity level was highest during the first 15 min of the confrontation and decreased over its course. For the unfamiliar goats, the activity pattern was similar but was modulated by rank, with higher values for low-ranking goats than for medium- and high-ranking ones. In conclusion, our results indicate that the presence of peers is advantageous for goats being introduced into groups of unfamiliar goats.