Hello! Our organization is looking into developing a stored procedure to match and merge some of our duplicates automatically. Have any of you ever done this before, or do you know anyone who has? I'd love to hear what worked and what didn't, and find an existing procedure to base ours off of.
We had a duplicate auto merge at SDC which was great.
I'm quite passionate about this topic. There is a real business case in labour saving and inaccuracy of inactive accounts vs the fear of incorrect merges deleting whole relationships.
For me the root problem is the difficulty in demerging.
I don't understand the binary of the problem. In terms of duplicates there are obvious through the various stages of likely, to unsure. If you can find obvious then merge them. The likelys then curate. The unsure, leave or survey for more data. The trick is testing your theories to prove yourself wrong (like any good scientist)