Employee voice mechanisms in organisations are designed to allow employees to have a say about their work. As new digital mechanisms, such as “social media” (SM), are increasingly adopted by organisations for knowledge-sharing, employee engagement and general communication, it is important to consider how SM may facilitate employee voice. The use of SM for voice is becoming more prevalent with noteworthy cases in the news, such as Google employees organising themselves over SM to protest organisational policies. This topic is even more salient as the increase of work-from-home (WFH) during the COVID-19 pandemic has accelerated the adoption of all types of SM for work communication. However, there is limited empirical evidence exploring the state of voice on SM and voice scholarship must examine SM’s potential as a voice mechanism, especially in the new normal. Employee voice ranges from organisation-related suggestions, concerns, and ideas to employee-related issues, grievances, or interests. Apart from a few recent studies, extant literature does not consider the nature of the SM mechanism and instead focuses on contextual or individual factors to explain voice on SM. This research adopts the technological affordance lens to understand how employees perceive the utility of different SM features, also referred to as SM affordances for different voice content. In addition, to the best of the author’s knowledge, current research examines SM in isolation and generally does not consider the role of other voice/communication mechanisms operating in the organisational voice system. This research uses the Voice Embeddedness Framework to compare SM to other mechanisms to understand the role of SM in an organisational voice system. In doing so, the research highlights the extent to which SM provides favourable voice outcomes for employees, such as increased participation in decision-making, or employees’ ability to make desired changes (influence), or even having their voice be heard by relevant voice targets. An interpretivist research paradigm with a qualitative case study design was adopted where two multinational telecommunication subsidiaries (cases) operating in Pakistan were compared to understand if voice on SM was unique to a specific organisational context. Semi-structured interviews were conducted with 52 participants using Microsoft (MS) Teams with stratified non-probability purposive sampling to explore individual perceptions of SM for voice. The data collection took place during COVID-19 lockdowns, where employees’ access to most voice mechanisms was limited and they were forced to rely on SM for communication. This specific context (professionals working in the IT sector during COVID-19 lockdowns) provided an excellent test bed for exploring the value of SM as a voice mechanism.