Advanced Techniques for Online Advertisement Spend Optimisation by Bruno Trentini
Decision-making in online advertisement campaigns is a challenging task for campaign and brand managers. It is often affected by uncertainty, latent variables, carry-over effects, biases, or even the time & costs pressure that the business units suffer when managing a plethora of audiences with various creative assets. Descriptive statistics and dashboards are part of the classic Business Intelligence toolkit, but dynamic environments where performance is driven by spend allocation require more powerful tools to infer optimal scenarios.
“How can we improve budget allocation across campaigns?”.
Join us and see how our team shed a light on the optimality of budget allocation problem with a Reinforcement Learning system that yields recommendations – a budget split that increases likelihood of maximum returns – to assist in the decision-making process. In this session, we will talk about the implementation and the technical components of this system.
Broadly speaking, we will chat about Bayesian statistics, reinforcement learning, Thompson sampling and optimisation.
Supported by
