Content re-monetisation: How to have your cake and eat it too

Abstract

The remonetisation of content is gaining momentum as (even the most profitable) content companies find themselves challenged for new growth opportunities and revenue streams. Remonetisation refers to the innovative use of media segmentation to repurpose long-form content (and associated ads) into forms suitable for new devices, storytelling and attention models, and new consumption styles. Media segmentation has hitherto been a complex trade-off between human intelligence and automated algorithms. Teams of human editors are cognitively sophisticated but slow(er) and less available. Automated algorithms provide speed, but only on tasks that machines do well. This paper proposes a model for cost-effectively repurposing content by triaging algorithmic innovation in media analytics while making novel use of crowdsourcing to provide a solution that acts as a force amplifier on both human and machine dimensions. We share a proof point of our remonetisation framework in our collaboration with significant sports network with complex and demanding needs on the segmentation vocabulary and system performance.