Export LinkedIn connections to CSV and import them into a CRM

You can export your first degree LinkedIn connections yourself: open Settings and Privacy, then Data privacy, choose Get a copy of your data, pick the larger archive that includes connections, request it, and LinkedIn emails a download link. The file will not hold everyone's email address, because LinkedIn says you only see addresses of connections who allowed it. Clean the CSV, remove people you should not contact, then import it into your CRM or into Sellinger.

Your connections are the warmest list you own. They already know your name, and many of them have never been written to by you since the day you connected. This post shows the official way to get them out, what to expect in the file, how to clean it, and how to import it. The steps and limits come from LinkedIn's own help page, read on 25 October 2026. LinkedIn changes its menus, so check that page if a label looks different.

How do you export your LinkedIn connections?

LinkedIn documents one route, the account data export. It does not scrape anything and it is allowed because it is your own data.

1. Click the Me icon on your LinkedIn homepage.
2. Select Settings & Privacy.
3. Click Data privacy in the left pane.
4. Under How LinkedIn uses your data, click Get a copy of your data.
5. Select the larger archive option that includes connections.
6. Click Request archive.
7. LinkedIn emails a download link to your primary email address.

The help page does not say how long the archive takes, only that you will receive an email. Plan for it not to be instant, and request it before you need it. If you manage several LinkedIn accounts, repeat the steps in each one, because each account holds its own connections.

What will the file contain, and what will it not?

LinkedIn's page does not list the columns of the connections file. In the files we have seen, the usual columns are first name, last name, profile URL, email address, company, position and the date you connected. That is an observation, not a LinkedIn promise, so open your own file and check before you plan a mapping.

LinkedIn does state three limits:

The first limit matters most. A profile URL is usually the one identifier that every row has, so plan to match and enrich on the URL rather than on email. If you need a way to fill the gaps, our post on finding an email address from a LinkedIn profile covers the options and their limits.

How do you clean the CSV?

Do this in a spreadsheet before anything touches a CRM. It takes less than an hour for most lists, and every minute saves a bad send later.

1. Open it as text first. Some exports start with a few explanatory lines above the header. Delete them so row one is the header.
2. Keep one row per person. Sort by profile URL and remove exact duplicates.
3. Normalise the URL. Trim spaces, remove tracking parameters after a question mark, and keep one format. The URL is your match key.
4. Split or combine names the way your CRM wants. Some CRMs take first and last name, others a full name.
5. Fix characters. If names look garbled, re-save as UTF-8. Sellinger repairs file encoding on import, but other tools may not.
6. Add a source column. Write "LinkedIn export 2026-10-25" on every row. Months later you will want to know where a contact came from.
7. Remove people you should not contact. Current customers, competitors, your own colleagues, friends, and anyone who asked you to stop. Put them in a suppression list instead of deleting the knowledge that they exist.

On the last step: being connected on LinkedIn is not consent to a sales sequence. Messaging rules differ by country and by channel, and none of this is legal advice. See our outreach rules and check the law that applies to you before a campaign.

How do you import the file into a CRM?

Every CRM with an import wizard works the same way: upload the file, map columns to fields, choose how to match existing records, and review a preview. Three choices decide whether it goes well.

• Matching. Match on email where you have one and on the LinkedIn URL where you do not. If your CRM has no field for a LinkedIn URL, create a custom field first, or the one identifier you can rely on is lost.
• Creating versus updating. Decide whether the import may only create new contacts or may also update existing ones. A safe first run only creates.
• Ownership and lifecycle. Set an owner and a lifecycle stage, or the new contacts will land with none and be invisible to the team.

Run a small test with 20 rows. Look at five records in the CRM, then import the rest. If you use HubSpot, Sellinger connects to it in both directions: you can import HubSpot contacts as leads and send leads, messages, meetings and deals back, as the HubSpot connection guide describes. Salesforce, Pipedrive and Zoho are coming soon and are not available yet. Our post on choosing a CRM for LinkedIn outreach compares what to look for.

How do you import the file into Sellinger?

Open Add Leads, then the CSV tab, and follow the three steps in the CSV import help: upload, map columns, confirm.

• Each row needs at least one of email, LinkedIn URL or phone. Your export will usually satisfy this through the URL.
• Files up to 50MB are accepted. Split anything larger.
• Extra columns, such as the connected date or your source column, can be mapped to custom fields.
• Leads already in your pipeline are skipped as duplicates and counted in the result, and the import report lists failed rows so you can retry them.
• Choose the destination so the leads attach to a campaign. A lead needs an agent to be contacted.

Sellinger can also add leads from LinkedIn directly, so a file is optional. The file is the better route when you want to clean, tag and segment the list first, or when you want the same contacts in a CRM.

What should you say to people who already know you?

A cold sequence is the wrong tool. These people accepted your invitation, so write as someone they know.

• Lead with the reason you are writing now, such as a new offer, a change at your company or a question about their work.
• Reference where you connected if you remember it.
• Keep the first message short and make the ask small.
• Use approval mode for the first batch. A campaign can run in approval mode, where every message waits for a person to approve it, and a person can take over any conversation.

Here is a made up first message, for illustration only: "Hi Dana, we connected after the logistics meetup in the spring. I am now helping freight teams answer quote requests faster. Would it be useful if I sent you what we changed? Happy to drop it if not." It names the context, gives a reason and asks for little.

What does this not give you?

An export is a list of people who accepted you, not a list of people who need your offer. Fit still decides results. In our own outbound, according to the Sellinger case study, the list and the message were bigger levers than speed. Read a sample of 50 rows before you send, and cut anyone who is not your buyer.

Do not use browser extensions or scripts to pull connections or profile data out of LinkedIn. LinkedIn's User Agreement, section 8.2, effective 3 November 2025, prohibits using software, scripts, robots or other means to scrape or copy the Services. The data export above is the supported route. For the wider picture, see our post on LinkedIn lead generation.