This article discusses measurement issues that arise when evaluators attempt to assess the meaning and extent of integrated service delivery in human services programs. Service integration is a concept that is currently enjoying a major revival. Current social policy directions (especially on the part of the Ontario government) reflect a renewed will to implement integrated service delivery models in human service programs and organizations. Earlier work, from the mid 1970s through the early 1980s, on measuring this slippery aspect of organizational and program functioning is now being reexamined, revised, and updated. Lessons learned from evaluating an inter-ministerially sponsored children's' services program in Ontario are presented here. Evaluation issues explored in this article include recommended procedures for measuring program stakeholders' visions of "ideal" integration as well as their views of the extent of "real" service integration in functioning programs and on how and why to compare "ideal" to "real" human service integration.