Five sources. One customer. Merge/purge makes your data agree.

Your customer data lives in more places than you'd like — the store system, the web orders, the donor records, the list a partner sent over. Mail from all of them and the same person gets three copies; count from all of them and nobody knows how many customers you actually have.

Combining sources is more than stacking them

Merge/purge takes records from multiple sources and produces one clean master. The hard part isn't the merge — it's recognizing that "Robert Smith, 123 Oak St" and "Bob Smith, 123 Oak Street Apt 2" are the same person while "R. Smith" across town is not. That takes matching logic built for how real data actually disagrees, and we've been refining ours since 1994.

Purging also has to be done carefully, because duplicates often know different things. Delete the copy carelessly and you delete the only record that had the customer's email. That's why we offer group posting: when two matched records each carry something the other lacks, this optional step posts the best of both onto the surviving record — available on request for an additional fee, so you keep the information without keeping the duplicate.

What you get back

One master, ready to use: every duplicate resolved to a single surviving record, chosen by your rules. Add group posting and that survivor also carries the strongest data from each of its duplicates. The results come back documented — what matched, what survived, and what each source contributed — so you can check the decisions rather than take them on faith. Fewer wasted mailpieces, honest customer counts, and one version of the truth going forward.

For one-time projects and production work alike

Some clients come to us once, to fold a handful of sources into a single master. Others — resellers, mailers, production shops — run multi-source merges with us on a schedule, with match rules and output formats we've agreed on together. Either way it starts the same place: a conversation with a real person about your sources, your rules, and what the final data needs to do. We'll tell you what the job requires — and what it doesn't.

How it works

1. 1

   Start with a conversation

   We'll look at your sources together and agree on the rules — how strictly to match, which source wins when records disagree, and what the final master needs to contain. Merge/purge lives and dies on these decisions, so we make them with you, not for you.

2. 2

   We match across every source

   Records are compared across all your sources with matching logic we've been refining since 1994 — built to catch the duplicates that aren't spelled identically, not just the exact ones.

3. 3

   Duplicates resolve to a single record

   Each set of matches collapses to one surviving record, chosen by the source priority you set. Want that survivor to also inherit the best data from all its duplicates — the email from one, the phone from another? That's an optional service we call group posting, available on request for an additional fee.

4. 4

   One master comes back, documented

   You get a single deduplicated master ready to use, with reporting that shows what matched, what survived, and what each source contributed — so you can verify the decisions instead of taking them on faith.