<abstract> <p>Development of Environmental Product Declarations (EPD)s used for green marketing, specification, procurement, certification and green building rating systems are important for documenting and understanding product environmental performance. Considering such applications any misleading of stakeholders has serious legal ramifications. Various studies have highlighted EPD veracity depends mainly on the data quality of underpinning life cycle assessment (LCA). This paper compares data quality across polyester product case studies, literature surveys and EPDs. Life Cycle Inventory (LCI) and Life Cycle Impact Assessment (LCIA) results are presented and interpreted. Surveys show recycled polyester fibre results are most sensitive to melt spinning energy data which varies over a wide range. The case studies compare results from median, lower and upper energy use in melt spinning. The work highlights that, accurate, clear definitions and vocabulary is as vital for specific foreground process data as it is for generic background supply chain data. This is to avoid misconceptions and mismatched assumptions in respect of EPD data quality and incorrect acceptance of inadequate charting of all essential processes. If product-specific accurate data is inaccessible, EPD options include presenting impact assessment results from LCI of best and worst-case scenarios. This is preferable to legal risks of using junk data that misleads stakeholders in marketing. General recommendations are presented for LCA practitioners to improve EPD data quality and accuracy. These include using multiple data sources to avoid reliance on any single database. Data also needs to be verified by a third-party with industry expertise independent of the specific manufacturer. It recommends using suitable, comprehensive and specific product-related scenarios for data development in any EPD.</p> </abstract>