Importing contacts from Gmail in two minutes

Most CRMs die at the import step. Here is the short version: export from Gmail, clean three columns, map them once, and check for duplicates before you invite the team.

The BittaFlow team

4 min read

On this page
  1. Step 1 — Export from Gmail
  2. Step 2 — Keep five columns, delete the rest
  3. Step 3 — Fix the phone column first
  4. Step 4 — Map once, then save the mapping
  5. Step 5 — Decide the duplicate rule before importing
  6. Step 6 — Import a small batch first
  7. Step 7 — Import the rest, then stop
  8. A word on consent
  9. What to do with the messy records
  10. The twenty-record test

The hardest part of moving to a CRM is not the software. It is the first hour, when you have to get your contacts out of Gmail and into something else.

It is also where most migrations stop. A file gets exported, looks wrong, and the project waits for a quiet week that never comes.

This is the short version, in the order that avoids the usual mess.

Step 1 — Export from Gmail

In Gmail: Contacts → Export → choose All contacts → Google CSV → Export.

You get a file called contacts.csv. Open it in a spreadsheet, not in a text editor.

A note on Google CSV versus Outlook CSV: choose the Google format if you are importing into most modern CRMs. It keeps the labels you actually use.

Step 2 — Keep five columns, delete the rest

Most of the file is noise: notes Google added, groups, photo links, birthdays. Delete everything except:

  1. First name
  2. Last name
  3. Phone
  4. Email
  5. Notes (keep this — it often holds the only context you have)

If your CRM has a company or tag field, keep that too. Everything else can go. A smaller file imports faster and fails less.

Step 3 — Fix the phone column first

This is the step people skip, and it is the one that causes silent failures.

  • Pick one format and apply it to the whole column. Include the country code if you serve more than one country.
  • Delete anything that is not a number: spaces, dashes, brackets are usually fine, but text like “call after 5” belongs in notes.
  • Look for landlines mixed with mobiles. If you send SMS, those landlines will bounce.

A quick way to find problems: sort the column and look at the shortest and longest entries. The outliers are almost always wrong.

Step 4 — Map once, then save the mapping

Your CRM will ask which column is the first name, which is the phone, and so on. Do this carefully once and save the mapping if the tool allows it.

Why it matters: the next import — a trade show list, a referral sheet — will use the same mapping, and you will not have to think about it again.

Step 5 — Decide the duplicate rule before importing

Two records for the same person is worse than one missing record, because nobody trusts the list after that.

Choose a rule:

  • Phone as the primary key. Usually the most reliable for service businesses.
  • Email as the primary key. Better for B2B, weaker for households that share an address.
  • Both, with phone winning. A sensible default for most clinics, trades and salons.

Then decide what happens when two records collide: keep the newer, keep the one with notes, or merge the fields. Write it down. The importer will follow whatever you choose, including a bad choice.

Step 6 — Import a small batch first

Take twenty contacts, import them, and check them by hand:

  • Does the phone dial correctly?
  • Is the name in the right field?
  • Did the notes survive?
  • Did anything get flagged as a duplicate that should not have been?

Twenty records will expose every problem the full file would, without creating a mess you have to clean twice.

Step 7 — Import the rest, then stop

Do not send a campaign the same day. Do not invite the whole team until the data looks right.

A useful order:

  1. Import.
  2. Check duplicates.
  3. Tag the list by source, so you know where it came from.
  4. Invite one colleague, let them use it for a day.
  5. Then invite everyone.

Importing a contact is not the same as permission to message them.

If the list includes people who never asked to hear from you, keep them out of bulk messages. Importing for record-keeping is one thing; sending marketing to an old export is how a domain gets a reputation problem.

What to do with the messy records

There will be some: a name with no number, a number with no name, three variations of the same person.

Do not delete them to make the numbers look clean. Put them in a “needs review” tag and clear it over a few weeks. A record you cannot use today may be the only trace of a customer from two years ago.

The twenty-record test

BittaFlow imports a Google CSV directly, maps the columns, de-duplicates on the rule you choose, and keeps the source tag on every record, so you can see later where a contact came from. The whole flow is built to be done in one sitting, not run as a project.

If you have been putting the import off, do the twenty-record test today. Ten minutes, and you will know whether the rest will work.

Useful? Send it to someone who runs the phones.