The wood furniture industry, similar to the timber construction sector, is increasingly interested in replacing conventional fossil-based adhesives with more sustainable bio-based alternatives. A representative application in furniture manufacturing is bed slats, which are curved, load-bearing wooden components bonded with structural adhesives. Typical bed slats consist of a seven-layer unidirectional ply with a total thickness of approximately 10 mm, a span of 700–900 mm, and a curvature radius of about 5000 mm. A critical challenge for bio-based adhesives in such applications is achieving sufficient and reliable bond strength. Current adhesive development strategies largely rely on trial-and-error approaches using tensile lap-shear tests to link adhesive formulation, curing conditions, and application [1]. However, it is well established that such tests are not suitable for predicting joint performance at the prototype or structural scale. Recent research has characterised the failure behaviour of lignin-based bio-adhesives using fracture mechanics concepts, where crack initiation and growth are described by traction–separation (cohesive) laws [2]. Building on this work, the present study aims to develop a fracture mechanics-based experimental and numerical framework for adhesively bonded wood components that can be readily adopted by wood adhesive and wood product manufacturers. Prototype bending tests of bonded bed slats are currently under way, with crack initiation and propagation monitored using digital image correlation. In parallel, a finite element model is being developed in the commercial software Abaqus/CAE, implementing experimentally derived cohesive laws through cohesive zone modelling. Different modelling strategies are being evaluated, ranging from simplified fracture energy-based approaches to full cohesive law implementations. The ongoing work will assess which combinations of material characterisation methods, prototype testing metrics, and numerical modelling approaches provide acceptable predictive accuracy, with the goal of supporting the development of reliable bio-based adhesives for future wood product applications.