Macroalgal forests (formed by kelps and fucoids) are lost worldwide due to several cumulatiehuman impacts, resulting in substantial losses of ecosystem functions and services. Consequently, they are listed as 1 of the 7 habitat types to be restored in the context of the recently approved European Nature Restoration Law (NRL), together with i) seagrass beds, ii) shellfish beds, iii) maerl beds, iv) sponge, coral and coralligenous beds, v) vents and seeps and vi) soft sediments. Under the NRL, ecological restoration planning at the national level will require a clear definition of actions and target areas. As marine restoration ecology is still in its infancy, in particular concerning marine forests, it’s important to capitalize on previous experience and carefully assess the feasibility and likeliness of success of each action on a case-by-case basis.Human-driven proliferations of grazers are one of the main threats for marine forests, causing regime shifts that lead to forest loss and the spread of less complex communities, such as algal turfs and barren grounds. Recent experiments have shown that grazing pressure is also one of the major causes of restoration failure. Very efficient grazers in marine forests, such as sea urchins, can deplete forests over very short periods causing the shift to barren grounds. Recent research proved that native, invasive and range-expanding herbivorous fish can also deplete marine forests quickly but can alternatively principally impact reproductive potential (i.e. Sarpa salpa and Ericaria amentacea), with visible effects over larger (and unknown) time scales. The role of mesograzers is however less known: we show that mesograzers can actively graze early-life stages of canopy forming species (i.e.Idothea balthica and Gongolaria barbata), but as their distribution, abundance and trophic behaviour is still poorly studied, more research is needed to better understand their role in the development of young forests and for restoration action.Reasoned restorative actions rely on the knowledge of the environment, such as historical evidence of degradation/loss of forests and the state of the impact(s) recognized as the cause(s) of degradation (or the potential mitigation). Grazing pressure should always be considered when assessing the suitability of a potential restoration site, even when it’s not (or it has not been recognized as) the cause of forest degradation/loss. When grazers are abundant in a particular area, restoration action will not be successful, and the project should be abandoned unless regulation of herbivores is feasible.Restoration success depends also on technological knowledge (including mitigation measures) and a societal context which must be part of a continuum including success evaluation and monitoring over the long term. In the framework of past and present European projects (i.e. Afrimed and FORESCUE), a decision-support tool has been conceived in collaboration with stakeholders. Consultation with stakeholders is paramount and should include, among others, scientists, decision-makers (national, regional or local governments), managers (MPAs), commercial users (fishermen, touristic activities), restoration promoters (NGOs and consulting), educational representatives and if possible - citizens (citizen scientists) in view of the application of the NRL at the local scale.