Biorefineries offer a new solution for the sustainable production of chemicals, materials, fuels, heat, and power while reducing dependence on non-renewable resources and helping to mitigate climate change. To ensure a broad implementation of biorefineries, efficiency is vital in the planning phase and must be monitored during operation. However, optimizing biorefinery processes can be challenging due to the inherent complexity of biomass feedstock, its variability, and the several technology combinations available. Digital mapping or modeling through approaches such as Digital Twins can lead to a better understanding of the processes involved. A Digital Twin fed by continuous process data can contribute to optimizing biorefinery processes, improving efficiency and helping to minimize adverse impacts. However, acquiring all relevant process parameters can be challenging due to the need to implement sensors and the volume and quality of the generated data. Here, we present the concept of Lean Digital Twin for biorefinery optimization to achieve efficient operation and monitoring while keeping overall data acquisition, processing, and storage as simple as possible. We show that implementing data-driven optimization strategies can help ensure the sustainability of biorefineries.