Background & objectives: Due to the dangers of the construction industry, leading and lagging safety indicators have been developed to measure safety performance and prevent injury. It is important to examine the effect of leading indicators on a project level to better understand how leading indicator data can be used by company management. This study examines the relationship between safety leading and lagging indicators when measured on a company level using company administrative data. Methods: This case study collected safety indicators from 47 construction projects. Four zero-inflated Poisson models were run to determine whether an increased number of leading indicators, site inspections or toolbox talks, led to a lower frequency of lagging indicators, injuries or first aid injuries. Results: There were few injuries in the dataset across all projects. Findings from univariate models showed the expected relationship between higher site inspections and toolbox talks and lower injuries and first aid injuries, although these findings were only significant with first aid outcomes. The estimated effect sizes of these models were very small. Conclusion: Although these results parallel some past studies, the limited number of injuries common to most single employers prevents adequate data for statistical analysis. Population level studies with multiple employers will more likely have adequate power to show associations in safety metrics. Single employers may use their data as a benchmark and show trends over time. However, employers should closely examine the quality of their data, and collect relevant variables to track the progress of their safety metrics.