Health communication campaigns have been used to address many of the most prevalent non-communicable disease risk factors, such as physical inactivity. Typically, campaigns are shared via mass media to reach a high proportion of the population and at a low cost per head. However, the messages shared are in direct competition with other campaigns, such as product marketing, which can result in the campaign not being seen adequately to lead to behaviour change. Moreover, as health campaigns are shared widely, the messages may not be understood or considered appropriate by certain audiences due to their broad nature. This can lead to unintended consequences, such as inadvertent social norming of the risk behaviour. To improve the success of health communication campaigns, they should be based on theory, with the theory of planned behaviour, the elaboration likelihood model, and the extended parallel process model, three of the most widely used. Such theories highlight the importance of targetting a campaign to the audience. Targetting a health communication campaign involves considering the audience in the development and dissemination of the message. Campaigns could also be co-developed with the audience in question to ensure relevance. Digital technologies such as machine learning and artificial intelligence can be used to tailor messages to the target audience effectively. Examples of targetted and broad health communication campaigns are presented.