Your Next Job Is Probably One Weak Tie Away
Most networking advice sounds like punishment.
Go to events. Collect contacts. Post every day. Ask strangers for coffee. Perform professional enthusiasm until somebody gives you a referral.
The evidence points to something both easier and more specific: the most useful person in your job search is often someone you already know a little.
The causal LinkedIn experiment below was published in 2022 and remains the strongest large-scale test of weak ties. The outreach method is current for 2026, but adapt the channel, tone, language, and referral norms to your local market. In some countries, professional associations or alumni groups matter more than LinkedIn.
Not your best friend. Not a stranger. The former coworker you liked but have not spoken to in two years. The person from another team who once helped on a launch. The classmate whose career drifted into an adjacent industry.
Researchers call these people weak ties, and LinkedIn accidentally built one of the largest employment experiments ever run around them.
The 20-Million-Person Experiment
From 2015 through 2019, LinkedIn varied its "People You May Know" recommendations. Some users were shown more people with whom they had strong ties; others saw more weak ties. Researchers then followed what happened.
The experiment covered roughly 20 million users, 2 billion new connections, 70 million job applications, and 600,000 jobs attributed through LinkedIn.[1]
Weak ties caused more job mobility than strong ties. But the result was not "message the most distant stranger you can find." The highest-yield relationships were moderately weak: people just outside your daily circle, often with around ten mutual connections in the study's structural measure.[2]
That makes intuitive sense. Your close friends know many of the same people and opportunities you do. Total strangers have new information but no reason to spend social capital on you. Acquaintances sit in the useful middle: different information, enough context.
A weak tie is valuable because their information is less redundant—not because they are more powerful. The goal is to reach adjacent networks, not chase impressive titles.
Build a Reconnection List, Not a Contact List
Open your work history and list people from four groups:
- Former coworkers who moved to a different company or industry
- Clients, vendors, and agency partners you worked with directly
- Classmates or community members doing adjacent work
- People you met through one substantial project, event, or collaboration
Choose ten. That is your network for this month.
The message does not need a disguised ask:
Hey, I saw you moved into developer relations at Acme. I always liked working with you on the launch project. I'm exploring roles that sit between product and community and would love to hear what surprised you about the move. No rush—hope you're well.
It works because it is true, specific, and easy to answer. You are reopening a relationship, not sending an invoice for a referral.
Ask for Information Before Access
The worst first message is "Can you refer me?"
The better sequence is:
- Reconnect around shared context.
- Ask one question only they can answer.
- Listen for a real fit.
- Mention the kind of move you are considering.
- Let them offer a next step—or make a small, specific request after the conversation.
A referral from someone who barely knows your work is weak evidence. A message that helps you understand how a team hires, what problem it is solving, or which capability it values can improve every application you send.
That distinction matters because referrals are useful partly as information. A Bureau of Labor Statistics working paper found evidence that referrals from friends and family give employers information they would not otherwise have about an applicant.[3] Your job is to give the person enough real material to stand behind.
Weak Ties Need Maintenance
You cannot reactivate fifty dormant relationships the night before you need a job and expect it to feel natural.
The maintenance version is tiny:
- Congratulate people on work you genuinely understand
- Send an article with one sentence about why it reminded you of them
- Answer questions when you can
- Keep short notes about the last conversation
- Follow up when someone gives useful advice
LinkedIn's Economic Graph research defines network strength partly by information value, message frequency, and how non-redundant your connections are.[4] A large network of people who all know one another is not the same as a broad network that crosses functions, industries, and communities.
- ✓Pick two former collaborators outside my current company
- ✓Send one specific, no-pressure reconnection message
- ✓Ask about their work before asking about openings
- ✓Record the useful context and promised follow-up
- ✓Offer help or information when I genuinely can
The Part Everyone Gets Backward
Networking is not a separate performance you do on top of work. It is the residue of doing useful things with people and staying lightly connected afterward.
If your network feels thin, do not compensate by broadcasting harder. Work backward through the projects, teams, classes, communities, and clients already in your history. The bridges are probably there. They are just quiet.
Your strongest opportunity may come from someone you would never call a close contact—which is precisely why they can see a path you cannot.
- The Developer
Sources
Summary of the five-year LinkedIn experiment involving approximately 20 million users, 2 billion connections, 70 million applications, and 600,000 jobs.
A Causal Test of the Strength of Weak Ties—the peer-reviewed experiment showing that moderately weak ties produced the greatest job mobility.
Research on what employers learn from referrals and how different referral sources convey applicant information.
LinkedIn's network-strength framework, including information value, communication bandwidth, and non-redundancy.