Some religions have a tradition of going door to door in search of people who they can share the gospel with and eventually recruit as church members. Internally, this practice has become controversial in some denominations as privacy has become a stronger value in American society and knocking on anyone's door without an appointment has come to considered bad manners. (How can one represent Christ rightly when one starts by behaving badly?)
For many years there has been strong data showing that this method and others that attempt to share faith with strangers are very ineffective. For example, Jehovah's Witnesses (JW) are required to put in several hours each week going door to door, but 95 percent of them do not have any converts in any given year. Another example, a Fundamentalist Christian group gets one convert per each 7,619 hours (per individual member) using this approach. The Later-Day Saints (LDS or "Mormons") have modified their approach to missionary work in recent years in view of the fact that most religious conversions in any faith take place within the social networks of the members. Their missionaries still have to put in some time doing cold-call door to door contacts, but they are also using TV spots to get leads for visits and encouraging more attention to members sharing with their friends, relatives, coworkers, etc.
Rodney Stark (who seems to be positioning himself as the iconoclast of the sociology of religion) and Laurence Iannaccone presented a paper at the Society for the Scientific Study of Religion meeting in October in Louisville arguing that there may be a marginal usefulness in organized programs of witnessing to strangers. Their paper argues that if a religious movement lasts long enough, its growth rate declines and eventually there is "network exhaustion." That is, all of the people in the social networks of their members who are likely to convert have already done so, or at least so many have done so that it is difficult to find potential members. They useJW and LDS growth data to support their theory that some movements "transcend the strong-tie barrier and grow through weak ties." They also say that a religious movement has to have a hierarchy with sufficient power to push people to do the difficult work involved in sufficient volume to actually continue to grow.
There are many difficulties with this paper by Stark and Iannaccone. (1) A single paper does not trump scores of previous studies. (2) This paper does not attempt to refute the primary finding that the most effective means of evangelism is among the social networks of members, it simply offers a rationale for continuing the cold-call approach despite this reality. If you hear someone referring to this research as an argument for door to door work and against "Friendship Evangelism," they simply don't know what they are talking about. (3) The JW data is suspect. The JW organization does not report reliable data and so any researcher who uses JW data is either getting it from more difficult sources (such as surveys in which people identify a JW preference whether or not they are actually adherents) or simply guessing. (4) There is plenty of evidence that LDS is reaching into new communities and has not come near "network exhaustion," so they are probably not a good source of data to support the theory proposed.
Bottom Line: The most effective kind of evangelism is still the informal sharing and demonstrated compassion of believers among their friends, relatives, coworkers, neighbors and acquaintances. Religious leaders who persist in promoting cold-call methods may have a variety of reasons for doing so, but that doesn't make it a good idea. If a congregation or movement feels that it is nearing "network exhaustion," it would be more effective to encourage their members to widen their networks (make new friends, get involved in new circles) than to adopt door to door canvassing as a key strategy.