Asymmetric Leverage in Networking | Generated by AI
Question: What do you think about Reid Hoffman’s networking philosophy — helping with the single most critical step at minimum cost, rather than being an all-around people-pleaser?
Answer:
It’s correct, and it maps cleanly onto a deeper principle: asymmetric leverage. The same thinking that makes good code (do one thing well, minimal surface area) makes good networking.
A few things worth pulling apart:
Why “help fully” backfires
When you offer unlimited help, you create two problems: (1) diffuse commitment that’s hard to cash, and (2) you signal low value — abundant, unconstrained resources are cheap. The 5-minute favor works because of the constraint. Scarcity of your attention is what makes the signal credible.
The ROI framing is honest, not cynical
Hoffman frames networking as investment with returns. Some people find this cold. I think it’s just accurate — pretending relationships are “purely human” while unconsciously tracking favors is worse than being explicit about it. The honest frame lets you optimize consciously.
Where it maps onto your situation specifically
You have ~3-4 productive hours/day outside work and family. The 5-minute favor principle is actually an energy conservation strategy more than a networking one. You can’t afford deep social investments right now. But you can afford:
- One precise technical answer to someone in your network
- One intro email (2 sentences)
- One link with a “thought of you” annotation
These compound into reputation without burning your engineering time.
The one thing I’d push back on
The article frames weak ties as universally more valuable than strong ones for opportunity. That’s true for job leads and information. But for your trajectory specifically — training models, building agents, deep AI work — you need at least 2-3 strong technical ties: people who will tell you when your architecture is wrong, read your code critically, or co-found something. Weak ties get you in the room; strong ties make what happens in the room worth anything.
Practical translation for you
Your blog and notes at lzwjava.github.io are already doing the 5-minute favor at scale — one post that unblocks 50 people who find it via search. That’s higher ROI than most individual networking. Keep that, and be very selective about 1:1 time.
The summary: Hoffman is right that most people waste networking energy on breadth of obligation. The fix is precision, not volume.