In bacterial genetics, large-scale screening approaches such as sequencing transposon mutant pools can be highly effective for identifying and characterizing genes with unknown functions. In the plant pathogen, Xylella fastidiosa, this approach is challenging due to the fastidious nature of this bacterial species and its niche-specific growth in the plant xylem tissue. The purpose of this study was to explore the feasibility of transposon sequencing (Tnseq) for identification of virulence genes in X. fastidiosa, with the hypothesis that this would uncover genes or pathways not previously associated with plant infection. Predicted essential genes were compared after X. fastidiosa strain M23 was grown in vitro and in planta using two known susceptible host species (grapevine and almond). After growth in planta, several gene categories were predicted as essential including hemagglutinins, tRNAs, toxin-antitoxin systems, and prophage genes. Three predicted essential genes (XfasM23_0359, XfasM23_0360, XfasM23_0972) were chosen for further validation by making targeted deletion mutants. Deletion mutants exhibited reduced disease in grapevines, but normal growth and aggregation phenotypes in vitro. Overall, the Tnseq approach has some practical limitations due to the nature of the X. fastidiosa pathosystem, and significant bottleneck effects of inoculation, but was still able to identify some genes contributing to disease in plants. Recommendations for future Tnseq studies in X. fastidiosa are discussed based on the challenges and results of this work.