Rapport Contact Deduplicator

What it is for

Every contact list collects duplicates: someone fills in a web form with a different e-mail, a colleague adds a business card by hand, an import runs twice. Exact-match tools only find identical rows. This tool looks for the same person: the same e-mail even with other capital letters, a +tag or (for Gmail) dots; the same phone number written differently; or a very similar name at the same company. It groups the matches, explains each one and writes a merged CSV.

It runs the same matching as Edit → Find duplicates… in Rapport, where you choose the record to keep and every deal and activity moves along.

Worked example

The example is a HubSpot contacts export with nine rows and HubSpot's own column names:

Found 3 groups of duplicates in 9 rows (3 records to merge).

• Group 1 (rows 2 and 3): two Lena Vogt records with lena.vogt@mediq.example and Lena.Vogt@mediq.example — same e-mail; same name at the same company (Mediq Health GmbH and Mediq Health count as one company), score 1.
• Group 2 (rows 4 and 5): Paul Hansen with 040 2345678 and +49 40 2345678 — same phone number, score 0.9.
• Group 3 (rows 7 and 8): Sanne Bakker and Sanne Baker at Greenway Logistics B.V. and Greenway Logistics — similar name at the same company, score 0.75.

A hand-made list with six rows, Name, Email, Phone, Company as headers:

Found 2 groups of duplicates in 6 rows (2 records to merge).

an.na.schmidt@gmail.com and annaschmidt+shop@gmail.com are the same Gmail inbox, so the two Anna Schmidt rows form a group. Jonas Weber and Jonas Webber at Weber IT and Weber IT GmbH are matched by name. Tom Berg and Thomas Berg at Berg Bau are not matched: a nickname is too different for the name check, and their e-mails and phones differ.

How to read the result

• Rows / Groups / Records to merge — records to merge = rows that would disappear into another one.
• Duplicate groups — the CSV row numbers (the header is row 1), the records, the reasons and a score from 0 to 1; e-mail matches score highest, similar names lowest.
• Columns used — which of your columns the tool read as name, e-mail, phone, company and so on. Check it first when the result looks empty.
• Downloads — contacts-merged.csv keeps the first row of each group, fills its empty fields from the others, combines tags and notes and keeps the newest documented lawful basis; duplicate-groups.csv lists the groups for review.

Tips and common mistakes

• Look at every group before you delete anything in your CRM. Two real people can share an office phone number, and groups are joined: if A matches B and B matches C, all three end up together.
• Score 0.75 means "similar name at the same company" — the weakest kind of match. Check those by hand.
• Shared inboxes such as info@ or office@ count as the same e-mail, so two people who both wrote from info@ land in one group — split those by hand.
• Put the record you trust most first in the file: the merged CSV keeps the first row of each group.
• Free-mail domains (Gmail, Outlook …) do not count as a shared company for the name check.

FAQ

Does it catch typos in e-mail addresses?

No. anna@exmaple.com and anna@example.com are different addresses; only a matching name at the same company would group them.

Why is a phone match not a 100 % score?

Phones are shared more often than e-mails — switchboards, family landlines — so the tool trusts them a little less.

Which columns does it need?

A name (full or first and last), and e-mail, phone or company to compare by. Other columns go into the merged file when Rapport knows them.

Is my list stored?

No. It is processed for this run only.