The advice you have been given is to build a network.
It is bad advice, and the reason it is bad is arithmetic.
A network is a number you grow slowly and can never verify. One person who trusts you is a thing you already have, and probably several of them, and you have not spoken to any of them in fourteen months.
The line worth keeping
Be one person for someone, rather than someone for everyone.
Read that twice, because it inverts how most founders spend their outreach hours. Being someone for everyone is a broadcast strategy: posts, comments, connection requests, a slowly rising follower count. It feels like progress because the number moves.
Being one person for someone is narrower and far more uncomfortable. It means that when a specific person has a specific problem, your name is the one that arrives in their head. Not a name. The name.
You cannot be that for a thousand people. You can be it for twenty.
Where they already are
They are not on a list you have to build. They are sitting in archives you already own.
LinkedIn messages. Old Slack DMs from a company you left. Team chats. WhatsApp threads. Email you never deleted. Every past job, every client, every project.
Friends. Managers who liked you. Old bosses. Classmates. Colleagues who moved on.
The friction is low because they already know who you are and, more importantly, how you work. That second thing is what you are actually selling, and it is the thing a cold prospect cannot assess at any price.
Twenty names, sorted honestly
Write down twenty people who already know you. Against each one, mark two things: do they know you, and do they have the pain you solve.
That gives you four groups, and only one of them deserves a message this week.
Familiar and in pain. Message now. This is the whole list that matters.
Familiar, no pain. Do not pitch them. Stay visible. Content, a check-in, a useful forward. Pain arrives on its own schedule and you want to be in the room when it does.
Not familiar, in pain. This is cold outreach. It works, but it is harder and slower, and doing it first is why most founders conclude that outreach does not work.
Not familiar, no pain. Ignore. Genuinely ignore. This is the group most people spend most of their time on, because it is the largest and the least frightening.
One correction that matters more than it looks: go to the person nearest the problem, not the person nearest to you. If you sell AI automation, the friendly salesperson you had lunch with is not your lead. The DevOps manager you barely remember is.
The reconnection, and why most of them fail
Here is the flow that works.
Open about them. Genuinely. “How’s life going? I saw you’re doing [the new thing].”
Then let it be about them for a while. Not two lines. A while.
Only later: “I wanted to let you know I’ve started doing something new as well. I’m helping [type of company] solve [problem] using [method]. If you know anyone in the space looking for help, let me know.”
Notice the last clause. You are not asking them to hire you. You are asking whether they know anyone.
That distinction is doing almost all of the work. Asking someone to hire you puts them in a position where the honest answer is often no, and no is awkward, and awkwardness ends threads. Asking whether they know anyone gives them an easy, generous action that costs them nothing and makes them feel useful.
Sometimes the person who knows someone turns out to be the someone. They will tell you themselves. Let them arrive at it.
The number you actually need
Most of them will ignore you. Some will reply warmly and do nothing. A few will go quiet halfway through. That is not failure, that is the normal shape of it.
You need one yes.
Not a pipeline. Not a funnel. One person who says yes, and then one project, and on that project you make as many friends as you can, because those people will move to other companies and carry your name with them.
The course this comes from cites two warm contacts that generated over a million dollars and over seven hundred and fifty thousand respectively across a career. Treat those figures as the instructor’s claim rather than audited fact, because they are not sourced. But the mechanism they describe is not exotic: one relationship, maintained across a decade, compounding through the moves that person made.
Why this beats the alternative right now
There is a reason the warm route is worth more this year than it was five years ago.
Platform-wide cold email reply rates have fallen from around 8.5 percent in 2019 to roughly 5 percent in 2025 and about 3.43 percent in 2026, driven by inbox saturation, tighter spam enforcement, and a flood of low-effort AI-generated outreach.
The channel that gets harder every quarter is the one where nobody knows you. The channel that does not degrade is the one where someone already does.
What this does not solve
It does not scale. Twenty warm contacts is a finite resource, and if you burn all twenty in a month with a badly judged pitch, you have spent an asset you took ten years to accumulate.
It also does not work if you have nothing to say. Reconnecting with no clear offer produces a pleasant conversation and no outcome. Know what you help, whom you help, and how, before you open the thread.
And it is not a substitute for the other two pipelines. Warm network first, partnerships second, recruiters third. Build all three, so that when one goes quiet you are not starting from nothing.
Do this today
Open LinkedIn messages. Open your old work Slack if you still have access. Open WhatsApp.
Find twenty names. Mark each one: familiar, and in pain.
Message the ones in both columns. Ask about them first. End with whether they know anyone.
Then stop, and wait, because you only need one of them to say yes.
Sources
- Cold Email Response Rate 2026 Guide. On the decline from 5.1 percent to 3.43 percent and its drivers
- Cold Email Statistics Based on Sending Over 20M Cold Emails. Woodpecker, on longer-term response-rate trends
- Why Most Cold Email Campaigns Are Dying in 2026. On inbox saturation and AI-written outreach