The matching algorithm begins with the Constituent Id. We did this as many of our clients wanted to add a column called Constituent Id to external spreadsheets and pre match some of their major donors. After that we use a combination of name and address fields that we have identified and perfected over the years. Occasionally a duplicate will sneak through but they can always be merged later.
As EasyREimport was originally built as a consultant tool to automate client imports we estimate that we have easily run 10,000,000 constituent records through this matching algorithm. We found it easier to get the records imported and then use the Raiser’s Edge Duplicate Constituent Management Tool to manage any possible duplicates. Here is a document describing duplicate management within the Raiser’s Edge:
(Download PDF) Reviewing and managing duplicate records