Abstract
Parents of children with special educational needs and disabilities (SEND) increasingly seek information and support through online discussions, yet little is known about how architectural features of digital platforms shape the interactions and support that parents encounter. The paper makes two contributions: First, it develops and applies a transferable social listening framework that integrates manual qualitative collection with automated application programming interface capture, topic modelling and discourse network analysis; second, it demonstrates how platform architecture shapes the forms of interaction and support that emerge across online SEND communities. Data were collected over a 12-month period across three platforms: Mumsnet, Facebook and Instagram, yielding 3312 posts and 19,349 interactions. Topic modelling and clustering identified thematic groupings, while discourse network visualisation revealed distinctive patterns of interaction between the platforms. Facebook and Instagram were dominated by informational and promotional content, including news, services and job opportunities, with relatively limited discussion between users. In contrast, Mumsnet generated fewer posts but substantially richer engagement, characterised by reciprocal experience-sharing, personalised advice and emotional support regarding school placements, Education, Health and Care Plans and individual needs. These findings demonstrate that platform architecture, specifically anonymity, content formats, moderation models and algorithmic curation, shapes whether parents encounter primarily broadcast information or reciprocal peer support. They have implications for researchers, policymakers and platform designers concerned with the design of online support and SEND provision across the four nations of the United Kingdom.