The fusion of virus and endosome membranes is an essential early stage in influenza virus infection. The low pH-induced conformational change which promotes the fusogenic activity of the haemagglutinin (HA) is thus an attractive target as an antiviral strategy. The anti-influenza drug Arbidol is representative of a class of antivirals which inhibits HA-mediated membrane fusion by increasing the acid stability of the HA. In this study two series of indole derivatives structurally related to Arbidol were designed and synthesized to further probe the foundation of its antiviral activity and develop the basis for a structure–activity relationship (SAR). Ethyl 5-(hydroxymethyl)-1-methyl-2-(phenysulphanylmethyl)-1H-indole-3-carboxylate (15) was identified as one of the most potent inhibitors and more potent than Arbidol against certain subtypes of influenza A viruses. In particular, 15 exhibited a much greater affinity and preference for binding group 2 than group 1 HAs, and exerted a greater stabilising effect, in contrast to Arbidol. The results provide the basis for more detailed SAR studies of Arbidol binding to HA; however, the greater affinity for binding HA was not reflected in a comparable increase in antiviral activity of 15, apparently reflecting the complex nature of the antiviral activity of Arbidol and its derivatives.