The research is based on questions launched online towards top management of large global companies worldwide (revenue above $5 billion) by the end of 2020, including the two main waves of the covid-19 pandemic. The sample was stratified to be representative of. For additional test of sample representativeness, see . The analysis from which this research is based on is also unique in scope, and uses some of the most sophisticated data techniques available to date. The model on how corporate capabilities affect corporate revenue and profit dynamics, either directly or through amplification responses by firms during the pandemic has been identified thanks to a meta-analysis of the academic and management literature. The resilience drivers as well as the segmentation of resilient firms in the text comes from applying advanced machine learning techniques such as Random Forest and statistical clustering. Random Forest resilience prediction accuracy was more than 80%, and higher than prediction based on traditional regression techniques. Using parametric technique, each resilient driver and each factor within a cluster is statistically significant, with more than a 99/100 chance of being accurate. Finally, the base line model was tested for robustness on multiple dimensions, -industry versus all sample, profit recovery distribution shifted by +/-10%, removal of top 5% outliers, etc. Results remain qualitatively the same.
