Purpose - The paper aims to provide a theoretically informed critique of current measurement practices for word-of-mouth marketing (WOMM) campaigns. Design/methodology/approach - An exploratory field study is conducted on a real-life WOMM campaign. Data are collected from two generations of campaign participants using a custom-built Facebook app and subjected to social network analysis (SNA). We compare our theoretically informed measure of campaign reach with industry standard practice. Findings - Standard metrics for WOMM campaigns assume campaign reach equates to the number of campaign-related conversations. These metrics fail to allow for the possibility that some participants may be exposed multiple times to campaign-related messaging. In this exploratory field study, standard metrics overestimate campaign reach by 57.5 per cent. The campaign is also significantly less efficient in terms of cost-per-conversation. SNA shows that multiple exposures are associated with transitivity and tie strength. Multiple exposures mean that the total number of campaign-related conversations cannot be regarded as equivalent to the number of individuals reached. Research limitations/implications - SNA provides a sound theoretical foundation for the critique of current WOMM measurement practices. Two social-structural network attributes - transitivity and tie strength - inform our critique. A single WOMM campaign provides the field study context. Practical implications - The findings have significant implications for the development and deployment of WOMM effectiveness and efficiency metrics and are relevant to WOMM agencies, agency clients and the Word-of-Mouth Marketing Association. Originality/value - This is the largest field study of its kind having collected data on 5,000 WOMM campaign-related conversations. Participants specified precisely whom they spoke to about the campaign and the strength of that social tie. This is the first SNA-informed critique of standard WOMM campaign measurement practices and first quantification of offline multiple exposures to a WOMM campaign. We demonstrate how standard campaign metrics are based on the false assumption that word-of-mouth flows exclusively along intransitive ties.