Our automated scheduling solution aims to optimize the quality of linear schedules, while reducing the human effort needed for scheduling content during off-peak hours. Users are able to define several factors that determine what is important for them in the schedules they want to generate, for example they might want to minimize the cost of the content or they might want to maximize the predicted ratings. Our solution uses machine learning to improve over time.
Automated off-peak linear scheduling
Scheduling content on TV channels is a complex and time-consuming task, handled by experienced planning managers that have deep knowledge on how to attract and retain viewers by scheduling the content that is available to them in a way that optimizes their company’s mission.
There is a large variety of dimensions that make a schedule successful, and the relative importance of these dimensions differs between broadcasters. Our automated scheduling tool will allow the users to configure the prioritization of multiple dimensions, and it will leverage an advanced machine learning model that will evaluate and optimize how to continuously improve the interpretation of these dimensions. It is possible to define new dimensions that will be considered in the creation of the schedule, some measurements will be straightforward, data that is readily available in our scheduling system like costs and box office success, while others will be calculated by machine learning algorithms. We could for example predict the ratings of a transmission, or calculate the similarity of a programme with content that was previously successfully scheduled in comparable timeslots.
We focus on AI based scheduling of content during off-peak hours where automation can really make a difference. The expertise required for peak time scheduling is a totally different subject that involves a lot of strategic decisions, a business and marketing vision. This is currently beyond the scope of this tool.
Our software effortlessly integrates with the existing planning workflow. In a matter of seconds, the schedules will be optimally filled by considering multiple dimensions such as cost, ratings and stock optimization. When generating a schedule, users will be able to finetune which criteria they find most important. The AI algorithm will learn which choices result in the best schedule.
In order to help our customers capture their intended audiences and meet the expectations of their advertisers, MEDIAGENIX has developed a machine learning tool that can accurately predict the TV ratings of scheduled content for every relevant target segment. The predicted ratings are updated whenever new information (e.g. the schedule of competitors) becomes available. This information can be used to make changes to the schedule where required, and it will dramatically reduce the costs incurred due to mismatches between expected and actual ratings.
Linear broadcasters are looking to schedule off peak time at a lower cost while maintaining or even increasing the viewer success.
Especially scheduling during off-peak hours is (since recently) considered for automated scheduling. The broadcasters are looking at two techniques to reduce the cost of the off peak schedule. The first is to use the available stock better, the second technique is to reduce the human effort that is required to perform that scheduling.
This is exactly what our solution will enable them to do. We are using state of the art technologies like machine learning and serverless architectures, while being informed about the complexities of our customer’s daily operations, thanks to our 25+ years of experience in the broadcast industry.