Monitoring behaviour and activity can offer valuable insights into the health and welfare of animals, helping animal carers, researchers, and conservationists to make informed decisions about animal management. Traditional behavioural assessments have relied on time consuming and non-continuous visual assessments, driving a need to develop a more automated method for quantifying animal behaviour. While the automatic assessment of behaviour from video footage yield would be an ideal option in a zoo setting due to its low non-invasive nature, it often produces inconsistent results and requires a controlled environment. Wearable devices that directly measure animal movement yield better results. We have recently developed an accelerometer-based “smart collar” that, after training machine learning algorithms, can be used to remotely quantify the behaviour of any species large enough to wear the device. This presentation will showcase this technology and explore its potential applications in zoo and aquarium setting, as well as how it could be used to improve our understanding of endangered species in the wild.