# Import contacts from a CSV

> Prepare a CSV of contacts, fix rows in the import preview, run the import, and learn how existing contacts and organizations are matched.

## Before you start

- Anyone on your account can run an import; nothing is queued for approval.
- Imports create **contacts** (and, where needed, the organizations they belong to).
  There's no separate organization import.
- An import never overwrites a contact you already have. Read
  [How rows are matched](#how-rows-are-matched) before importing a list you've
  edited outside the portal.

## Steps

### Prepare the file

The file must be a `.csv` — the file picker won't offer anything else — with these
four column headers, spelled exactly like this:

| Column            | Required | Notes                                         |
| ----------------- | -------- | --------------------------------------------- |
| First Name        | No       | Stored as typed                               |
| Last Name         | No       | Stored as typed                               |
| Email             | Yes      | Lower-cased and trimmed before it's used      |
| Organization Name | No       | The organization the contact should belong to |

There's no column-mapping step, so a header that doesn't match is simply ignored and
its column is dropped. Only **Email** is required on a row; a row without a valid one
is marked invalid and skipped. Phone numbers, titles, addresses, tags, and industries
can't be imported — add those after the contacts land.

A file can hold up to **5,000 rows**. A bigger file is refused with "CSV exceeds the
maximum of 5000 rows", an empty one with "The CSV file is empty", and a file we can't
read with "Failed to parse CSV file". Split a longer list and run it twice.

Click **Import CSV** on the **Contacts** page, then **Download CSV template** to get
`import-template.csv`, a file with just the four headers to fill in.

_Screenshot: The Import Contacts dialog offering Upload CSV, Add Manually, and Download CSV template._

### Review the preview

Choose **Upload CSV** and pick your file — or **Add Manually**, which skips the file
entirely and drops you into an empty row with the cursor in **Email**. Either way you
land on the preview table, which is editable:

- Click any cell to edit it, then press Enter (or click away) to keep the change, or
  Escape to cancel.
- **Add Row** appends a blank row; the trash icon at the end of a row removes it.
- The badges above the table count **N valid**, **N invalid**, and **N duplicates
  skipped**, and update as you edit.
- Each row shows a status icon and, where something's wrong, the reason in the
  **Errors** column — "Email is required", "Invalid email format", or "Duplicate
  email".

_Screenshot: The Import Contacts preview step: a one-row editable table with First Name, Last Name, Email, Organization, and Errors columns, a 1 valid badge above it, and the Add Row, Back, and Import 1 Contact controls._

"Duplicates" here means **repeats within this file**: the second and later rows with
an email address that already appeared above. They're skipped so the same person isn't
processed twice. Rows matching a contact you _already have_ in the portal are not
flagged — they count as valid and are processed as described below.

Use **Back** to drop the file and start over. When the preview looks right, click
**Import N Contacts**; the button counts only the valid rows, and invalid rows are
skipped ("Invalid rows will be skipped during import.").

### Start the import

You'll see "Import started — you'll be notified when complete" and the dialog closes.
The import runs in the background in batches of 25 rows, so contacts appear in the
list gradually rather than all at once — refresh the page to see progress. You don't
have to stay on the page, and closing the tab doesn't stop it.

## How rows are matched

### Contacts you already have

If a row's email address already belongs to a contact on your account, that contact is
**kept as it is**. Its name, phone, title, and everything else stay exactly as they
were — an import is never an update. If the contact had been deleted, the same email
address **restores** it, with its old details and history intact.

This is why a 60-row import can report far fewer new contacts than rows, and why a
contact you deleted can reappear after someone imports an older spreadsheet.

### Organizations

When a row has an **Organization Name**, how we find the organization depends on the
email address:

- **A company email address** (anything that isn't a well-known personal domain) is
  matched by **domain**: we look for an organization whose **Domain** is the part
  after the `@`. If one exists, the contact joins it — _even if that organization has
  a different name than the one in your file_. If none exists, a new organization is
  created with the name from your file and that domain saved on it.
- **A personal email address** (Gmail, Outlook, and similar) is matched by **name**
  instead, ignoring capitalisation. If no organization has that name, one is created
  with no domain.

On top of that, every contact created anywhere in the portal — by import, by hand, or
through a form response — is linked to an organization whose **Domain** matches their
email address, even when no organization was named. Keeping **Domain** correct on your
organizations is what makes this work in your favour.

## What happens next

When the job finishes, two things arrive for the person who started the import:

- A **bell notification**: "Import Complete: N of M processed", with counts of the
  contacts and organizations created versus the ones that already existed.
- An **email** with the same summary, listing every row that failed and the reason it
  failed, so you can fix those rows and re-import just them.

Contacts created by an import behave like any other contact: they can be tagged,
edited, linked to opportunities, and included in response requests right away.

## Common questions

**Q: The import says 40 processed but only 12 contacts are new. Where did the rest
go?**
A: They already existed. Any row whose email address matches a contact on your
account is counted as processed but left untouched — imports never update an existing
contact, and repeats inside the same file are skipped as duplicates.

**Q: I updated names and phone numbers in my spreadsheet and re-imported. Nothing
changed.**
A: That's expected — an import only ever creates or restores contacts. To change
details on contacts you already have, edit them in the portal with **Actions** →
**Edit Contact**.

**Q: My contact ended up under an organization with a different name than my file
said.**
A: For company email addresses the organization is matched by email domain, not by
the name in your file. If an organization already has that domain, the contact joins
it. Fix it by editing the contact's **Organizations** field, or by correcting the
**Domain** on the organizations involved.

**Q: Can I import phone numbers, titles, tags, or addresses?**
A: No. The import reads only First Name, Last Name, Email, and Organization Name. Add
the rest afterwards on each contact's page.

**Q: Why were some rows skipped as duplicates when those people aren't in my account
yet?**
A: "Duplicate" refers to the file, not your account: the same email address appearing
more than once in the upload. Only the first of those rows is imported.

## Related

- [Contacts and organizations](https://kb.incenti.co/contacts-and-pipeline/contacts-and-organizations.md)
- [Request responses from contacts](https://kb.incenti.co/forms/request-responses-from-contacts.md)
- [Emails Incenti sends](https://kb.incenti.co/getting-started/emails-incenti-sends.md)
- [Notifications and the bell](https://kb.incenti.co/getting-started/notifications-and-the-bell.md)
