NuVentrec

Guides / Duplicate contacts

Guide · Moving your data

Duplicate contacts after an import: how to find them and merge them safely

Every real estate team that imports a list, joins two systems or moves CRMs ends up with the same seller twice: once from a skip-traced list, once from a web form, once from the old CRM. Duplicates split a conversation in two, make two callers work the same lead, and can send a second text to someone who already said stop. This guide shows how to find duplicates you can trust, how to merge without losing history, and how to stop new ones from appearing.

By the NuVentrec team · Published October 14, 2026 · Updated October 19, 2026 · Sources last checked October 19, 2026

The short answer

  • Normalize before you compare. Phone numbers in one format, emails in lower case, addresses in the Postal Service's standard form.
  • Match on strong keys first: the old system's ID, then phone, then email. Use name and address only as supporting evidence.
  • Never merge automatically on a weak match. Put those pairs in a review list for a person.
  • Merge, do not delete. Move every text, call, note and deal onto the record you keep, and archive the other one with a pointer to it.
  • An opt-out on either record wins. If one copy said stop, the merged person said stop.

Why duplicates appear

Duplicates are rarely careless typing. They come from systems that describe the same person differently:

  • Phone formats. The same number written with brackets and a dash, with only dashes, or as +1 and ten digits is still one number, but a plain text comparison sees three.
  • Address formats. "123 County Road 45" and "123 CR 45" are the same place.
  • Several sources. A list import, a web form and a referral each create a person before anyone checks.
  • Two systems in parallel. During a move, the old and the new CRM both receive new leads.

Step 1: Normalize the fields you match on

Comparison only works when both sides are written the same way.

  • Phone numbers: store them in the international E.164 form defined by the ITU's numbering plan: a plus sign, the country code and the number, with no spaces or punctuation. For US numbers that is +1 followed by ten digits.
  • Email: trim spaces and compare in lower case.
  • Addresses: the US Postal Service's Publication 28 sets the standard. It defines a standardized address as one that "includes all required address elements", using approved abbreviations for street suffixes and unit designators. Convert both sides to that form before comparing.
  • Names: trim, fix capital letters, and keep nicknames as they are. "Bob" and "Robert" are a hint, not a match.

Keep the original value next to the normalized one. You compare on the clean version and show people what they typed.

Step 2: Match in order of strength

  1. The same outside ID. If two records carry the same ID from the old system, they are the same person. Our guide to switching CRMs without losing history explains why every imported record should keep that ID.
  2. The same phone number. Strong, but not perfect: phone numbers get disconnected and reassigned to new people. The Reassigned Numbers Database exists because "numbers that were reassigned from a previous consumer" are a real problem for callers.
  3. The same email. Strong for individuals, weak for shared inboxes like an office address.
  4. Name plus address. Supporting evidence only. Families share an address; an heir and the late owner share a name and a property.

Pairs that match on step 1 can be merged with a quick check. Pairs that match only on steps 2 to 4 go to a person.

Step 3: Review the uncertain pairs

For each pair, show the two records side by side: names, numbers, addresses, the last few messages, open deals and the opt-out status of each. Most pairs take a few seconds to decide. Three outcomes are possible:

  • Same person: merge.
  • Different people: mark the pair as "not a duplicate" so it does not come back.
  • Not sure: leave both, and add a note to call and confirm.

Step 4: Merge without losing anything

A merge should be boring and complete:

  • Pick the record to keep, usually the one with the most history or the open deal.
  • Move everything onto it: texts, calls, recordings, notes, tasks, deals, properties and documents, each with its original time.
  • Fill empty fields from the other record, and keep conflicting values (two phone numbers, two emails) rather than choosing one.
  • Carry consent and opt-outs. If either record is on the do-not-contact list, the merged person is. Federal rules require a revoked consent to be honored within a reasonable time of no more than ten business days; a duplicate that "forgets" an opt-out breaks that rule. Our opt-out list guide covers the details.
  • Archive the other record, do not delete it. Keep a pointer to the record it was merged into, so old links and reports still lead somewhere and a mistake can be undone.

Step 5: Stop new duplicates at the door

  • When a form comes in, look the person up by phone and email before creating a new record.
  • When importing a list, run the same matching first and show how many rows are new, how many match an existing person and how many need review.
  • Run the duplicate check again after every large import, not once a year.

Our move checker reads an export in your browser and shows what it holds before anything is imported.

Duplicate deals and cards

People are not the only thing that doubles. After a move, the same property can sit on two pipeline boards: once from the old system and once from a list import. Treat cards like people: match on the old system's ID first, then on the person and the property address, keep the card with the most recent stage and history, and archive the other with a pointer. Do not merge two deals just because the address matches: a seller may own several lots on the same road.

How this fits in NuVentrec today

NuVentrec stores phone numbers in one normalized form and keeps a cleaned copy of each address for matching. Every import keeps the outside ID and starts with a check that shows what is new and what matches a person you already have. The Duplicates screen lists likely pairs for a person to review; a merge moves the history onto the kept contact and archives the other with a pointer to it, and opt-outs always carry over. See what is live or start now. Moving a spreadsheet in? Follow spreadsheet to CRM in one day, which brings opt-outs over first. Plans and what each includes are on the pricing page.

Questions people ask

Can we merge everything with the same phone number automatically?

It is tempting but risky. Reassigned numbers, shared family phones and office lines all produce false matches. Merge automatically only on the old system's ID and send the rest to review.

What happens to two different phone numbers when we merge?

Keep both on the merged person and mark which one is the main number. Throwing one away can lose the number the person actually answers.

We merged the wrong two people. Can we undo it?

If the other record was archived rather than deleted, yes: restore it and move its own history back. That is the main reason to never delete during a merge.

How often should we look for duplicates?

After every import or list upload, and once a month for the rest. The longer duplicates sit, the more history gets split between them.

Sources

  1. ITU-T Recommendation E.164, "The international public telecommunication numbering plan"
  2. US Postal Service, Publication 28, "Postal Addressing Standards", section 211, October 2024
  3. Reassigned Numbers Database (lookup of permanently disconnected US numbers)
  4. 47 CFR 64.1200, "Delivery restrictions", paragraph (a)(10), Legal Information Institute

One person, one record, every conversation in one place.

Try the interactive demoBook a walkthrough
Call now