The COVID19 pandemic profoundly disrupted global tourism, exposing structural vulnerabilities in destinations worldwide. This paper examines how pre-crisis customer base and geotourism diversity influenced the recovery of tourism in France and its neighbouring regions (Spain, Italy, Germany, Belgium, and the Netherlands). Using a modified tourism recovery indicator and diversification indices applied to monthly regional data from 2020 to 2022, our findings reveal that destinations with a higher share of domestic tourists prior to the pandemic recovered more quickly than those relying heavily on international visitors. Furthermore, regions exhibiting greater geotourism diversity, measured through the Shannon-Wiener and Pielou indices, experienced stronger recovery trajectories, particularly in attracting foreign tourists. This spatial diversification allowed tourism activity to adapt to mobility restrictions and shifting traveller preferences, mitigating losses in metropolitan areas through increased activity in rural and urban regions. We observe that domestic tourism acted as a stabilizing force during the crisis, while balanced territorial dispersion of tourist flows enhanced adaptability to shocks. Government restrictions and population density also played a significant role in shaping recovery patterns, underscoring the vulnerability of highly concentrated tourism systems. Neighbouring tourism and proximity travel also emerged as important drivers of resilience, helping destinations reduce seasonality effects and sustain activity during uncertainty. This research contributes to the literature on regional resilience by integrating market and spatial diversification as dual pillars of shock mitigation. Policy implications emphasize the need for differentiated regional strategies that foster diversified demand structures and balanced distribution of attractions. Strengthening these dimensions enhances long-term resilience and competitiveness, ensuring that destinations are better equipped to withstand future crises.