How to Export LinkedIn Connections and Combine Them Across Your Team

Every staffing firm has a version of this problem. Twelve recruiters, a few thousand LinkedIn connections each, and no way to see across them. A rep cold-emails a VP of Finance that the person two desks over used to work with. Nobody finds out until the deal is lost.

We have spent the last year building tooling for exactly this, and along the way we did the manual version more times than I would like to admit. For our own team, for design partners, and for three people who let us pool their networks into one firm's map. That last one took the firm from 3,767 reachable people to 6,682 in an afternoon.

Here is the manual version, with everything that bit us.

Step 1: Export your own connections

1. On LinkedIn, click the Me icon at the top right, then Settings & Privacy.
2. Click Data privacy in the left menu.
3. Under "How LinkedIn uses your data", click Get a copy of your data.
4. Choose Download larger data archive, including connections, verifications, contacts, account history... Do not hunt for Connections in the "Want something in particular?" list. It is not there anymore.
5. Click Request archive. LinkedIn emails a download link to your primary address. Sometimes it takes ten minutes. Usually it is the next morning.
6. Download the ZIP. There are a couple dozen files in there about you. The one you want is Connections.csv.

Open the file once and look at it. Count the rows against the number on your profile. It will be lower. More on why further down.

Step 2: Have every teammate export theirs

Who to ask: sales, recruiters, leadership, and the person who has been in the industry twenty years and "doesn't really use LinkedIn." That last one usually has the best list in the building.

Do it on one team call. It takes ten minutes for everyone to request their archive. A day later, each person forwards the ZIP (or just Connections.csv) to one owner, who renames each file with the sender's name.

Ground rules, in plain terms:

• Each person exports their own data, using LinkedIn's own feature. That is what the feature exists for.
• Nobody shares a login. Nobody installs a plug-in that pulls other people's connection lists.
• What you are building is an internal list for your own team. Treat it the way you treat the CRM.

One more thing to head off now: if someone offers you a Sales Navigator lead list instead of an export, it will not join to anything. More on that below.

Step 3: Stack the files into one sheet

Google Sheets. Make one tab per person and paste in their CSV with the note lines removed. In each tab, insert a new column A called Owner and fill it with that person's name (type it in A2, select down to the last row, Ctrl+D). Your columns are now Owner, First Name, Last Name, URL, Email Address, Company, Position, Connected On, in A through H.

Then add a tab called Combined and put this in A1:

=QUERY({Chris!A2:H; Dana!A2:H; Priya!A2:H}, "select * where Col4 is not null", 0)

Col4 is the URL column, so blank rows fall out. Add a tab, add its range to the list, and the combined sheet updates itself.

Excel. Put all the CSVs in one folder. Then Data, Get Data, From File, From Folder, pick the folder, and Combine & Transform. Power Query stacks the files for you and adds a Source.Name column, which is your owner column. In the sample-file step, use Remove Rows, Remove Top Rows, 3, so the note lines do not become data.

Step 4: Count who knows each person, matching on the URL

The instinct is to de-duplicate by name. Do not. Names collide (you have more than one Chris Smith in there), people write them differently in different places (we matched a "JB Hruska" and a "Jon Hruska" as the same person only because of the ID behind the name), and a name tells you nothing about which teammate holds the relationship.

Use the URL column. It is the linkedin.com/in/... address, and it is the closest thing to a person ID that the export gives you.

In Combined, add a column I with the header Known by, and in I2:

=COUNTIF($D$2:$D, D2)

Fill it down. Now every row says how many of your teammates are connected to that person.

If you want one row per person instead of one row per relationship, add a People tab. In A2, =UNIQUE(Combined!D2:D) gives you every distinct URL. In B2, this lists who knows them:

=TEXTJOIN(", ", TRUE, UNIQUE(FILTER(Combined!A:A, Combined!D:D=A2)))

Pull the name and company across with XLOOKUP against column D, and fill down.

Keep both tabs. Do not delete the duplicates. The duplicates are the point. A person who shows up in three exports is a person three of you can reach, and the Known by number is your first rough measure of how warm a path is.

One honest caveat: people change their profile URL. When we re-pulled one 2,466-person network a month later, 7 people had changed theirs and looked like new people. Small, but real. Do not expect a perfect count.

Step 5: Make it answer "who do we know at ___?"

This is the question anyone will actually ask the sheet, and the Company column is not ready for it. It is whatever the person's profile says today, spelled however they spelled it. "Acme Inc", "ACME", and "Acme Corporation" are three different companies to a filter.

Add a cleaned company column. In Google Sheets, with Company in F:

=TRIM(LOWER(REGEXREPLACE(F2, "(?i)[,.]|\b(inc|llc|corp|corporation|ltd|limited|co)\b", "")))

In Excel, Find and Replace does the same job in a few passes.

Then two things:

• A pivot table with the cleaned company as rows, a count of unique URLs as the first value, and a count of unique owners as the second. Sort by the first. That is your team's network by account, biggest first. It is usually a surprise.
• A filter view on Combined so anyone can type a company name and see the people, their titles, and which teammate is connected to each one.

Now run your target list through it. Take the twenty accounts you most want to be in this quarter, filter for each one, and for every hit the owner sends a two-line note that same week. Not a pitch. "Saw you're at Acme now, would love to hear what you're working on." That is the whole play, and you can run it Friday.

What the export will not tell you

This is the part most guides skip. Everything below cost us time to learn.

It is smaller than your profile count. Closed and restricted accounts still count in the header. When we pulled one 2,510-connection network through LinkedIn's own interface, 45 came back as "this profile can't be accessed," most of them connections from 2012 to 2018. Do not chase the gap. It is not there.

Emails are mostly blank. LinkedIn only includes an address when that person has allowed their connections to download it. Most have not. If your plan was "export everyone and load them into a sequencer," this is where it ends.

It only knows where people work today. No work history. That matters more than it sounds, because a lot of the best paths into an account are people who used to work there. We checked 58 senior contacts on one target list against their live profiles and 22 had changed jobs since the record was written. A CSV cannot tell you that, and a spreadsheet starts going stale the day you build it.

Connected On is not strength. Someone you connected with in 2014 and never spoke to sits in the sheet right next to your best client. The fix is manual and worth it: each owner goes through their own list once and marks the people they would actually message tomorrow. Yes, maybe, no. It takes about an hour per person and it is the most valuable column in the whole workbook.

Sales Navigator lists do not join. A Sales Nav lead list gives you URLs like linkedin.com/sales/lead/ACwAA..., which is a different ID for the same human than the /in/ address in the export. As text they never match. Match those rows on name plus company by hand, or ask the person for the real export.

It is first degree only. The export will not show you who your connections know. If you want second-degree reach, each of those people has to run the same steps.

Where the spreadsheet stops

Be honest with yourself about what you just built. It is a snapshot. Nobody will maintain it. The next three hires' networks will never make it in. It cannot tell you when one of those 6,000 people takes a job at an account you are chasing, and that move is exactly the moment a warm path is worth the most.

That is the manual version of what WarmPath does continuously. Each teammate connects their own LinkedIn (their own login, nothing shared), the firm gets one pooled map with work history behind every name, and you get told when a connection lands somewhere you are trying to get into. See the three-step framework for how firms put that map to work, or how it works for the product side.

But build the spreadsheet first. Even the rough version changes how your team thinks about the accounts on the board.

About the author

Evan O'Connor

Co-Founder, GTM at WarmPath

Evan O'Connor is co-founder of WarmPath, where he leads go-to-market. Before WarmPath he ran sales leadership at IntelAgree, a contract-AI startup, where he built the original 'knighting strategy' of warm-introduction prospecting that the WarmPath product is based on. He writes about warm introductions, relationship intelligence, and how staffing firms turn placements into pipeline.

LinkedInMore about WarmPath