There is great potential in public health and faith communities partnering to promote health education and research. This article describes lessons learned from the design and implementation of such a partnership, the Los Angeles Mammography Promotion in Churches Program (LAMP). It is feasible, although challenging, to enumerate and survey churches in a geographically defined urban population, using data compiled from telephone directories, religious sources (e.g., denominational networks), chambers of commerce, and newspapers. Among those sources, telephone directories provided the broadest coverage, whereas religious sources yielded the highest recruitment rate. For collecting survey data from churches, telephone methods are superior to mail methods. For church-based health education programs that seek comprehensive coverage and have adequate resources, it is recommended that telephone directories be combined with religious sources to identify target churches. For programs with limited resources and less emphasis on comprehensive coverage, using religious sources alone is recommended. (Contains 3 tables.)