Here's the thing about duplicate records: No matter what you do, you'll never get rid of them completely, because humans are human. Whether it's a staff member creating a duplicate record because they forgot to check if the record exists, or it's a customer who doesn't want to take the time to reset his password and just creates another record with a different email address, duplicates are going to happen.

And because duplicates are inevitable, you have to have a plan for dealing with them. Here are some suggestions:

1. Create data integrity reports that will help you identify potential duplicate records, and run those reports consistently. ("Consistently" means at least once a month, if not more often.) And make sure someone (or several someones) on staff has responsibility for cleaning up this data.
2. Make it everyone's job on staff to identify potential duplicates whenever they are using the database and make sure you have a clear process for how those duplicate records are reported. And make sure someone (or several someones) on staff has responsibility for cleaning up this data.
3. And did I mention to make sure someone (or several someones) on staff has responsibility for cleaning up this data?

Managing duplicate records is a journey, not a destination. No matter how good your technology is, like weeds in a garden, duplicate records are going to appear. So you have to have a plan for dealing with them, not just now, but always.