Comprehensive research metadata greatly clarify the methods and processes used to capture data and produce datasets. In so doing, they provide data users the information needed to better analyze, interpret, preserve and share them. Richly structured metadata are even more critical with complex longitudinal studies that contain thousands of variables, different data types and many waves of data collection. Midlife in the United States (MIDUS) is a national longitudinal study of approximately 12,000 Americans that examines aging as an integrated biopsychosocial process. MIDUS has a broad and unique blend of survey, experimental and laboratory research data collected over 20 years through a variety of modes. For the last decade, MIDUS has increasingly relied on the DDI to manage these complex research data and make their metadata machine-actionable in a userfriendly format. More recently, MIDUS has used Colectica tools to improve its DDI infrastructure and create a DDI-based harmonized data extraction system. Such a system allows researchers to search across datasets for variables of interest, identify and harmonize related longitudinal versions of variables, and easily create customized data exports.