Essential genes are the central hubs of cellular networks. Despite their importance, the lack of high-throughput methods for titrating their expression has limited our understanding of the fitness landscapes against which essential gene expression levels are optimized. We developed a modified CRISPRi system leveraging the predictable reduction in efficacy of imperfectly matched sgRNAs to generate specific levels of CRISPRi activity and demonstrate its broad applicability in bacteria. Using libraries of mismatched sgRNAs, we characterized the expression-fitness relationships of essential genes in Escherichia coli and Bacillus subtilis. Remarkably, these relationships co-vary by pathway and are predominantly conserved between E. coli and B. subtilis despite ~ 2 billion years of evolutionary separation, suggesting that deeply conserved tradeoffs underlie bacterial homeostasis.One Sentence Summary: Bacterial essential genes have varying responses to CRISPRi knockdown that are largely conserved across ~2 billion years of evolution.