Mesh

A 2022 study tested weak ties theory on 20 million LinkedIn users over five years. The finding: your weakest connections aren't always most valuable. Neither are your strongest. It's the moderately weak ties that create the most job mobility. 🧵⤵️ The experiment tracked 2 billion new connections and 600,000 new jobs. Researchers varied how often LinkedIn suggested weak versus strong ties. Result: an inverted U-shaped curve. Ties that were too weak provided diminishing returns. Ties that were too strong didn't help much either. Moderately weak ties—people you know but don't talk to often—had the highest impact. They're familiar enough to trust and reach out to, but distant enough to have access to different information and opportunities than the people you see every day. The effect varied by industry. In digital industries, weaker ties helped more. In traditional industries, slightly stronger ties performed better. But across the board, the extremes underperformed the middle. This is why Mesh tracks recency and context alongside basic contact info. Who you haven't talked to in 6 months. What you last discussed. Why you connected 2 years ago. Get started at me.sh! And read the full study at https://lnkd.in/gQJ78qEY

We're all wildly overconfident about what we'll remember from a conversation a month later. The favorite dish. The podcast they couldn't stop talking about. The thing they were nervous about. In the moment it feels unforgettable. Four weeks on, it's gone. So one of the oldest ideas in Mesh is also the simplest: when in doubt, write it down. A few things we've learned about doing it well: • Notes live on the person, not in a pile. Every note you take attaches to whoever it's about and stacks up in order, so their history is right there next to their name. • There's no character limit, so don't self-edit. You never know which detail turns out to matter. • Write down what someone needs help with, even if you can't help right now. When they resurface, you'll remember—and sometimes by then you can. • A note written for a person is a gift to your future self, three months from now, walking into a coffee you'd otherwise have shown up to empty-handed. Notes, done right: https://lnkd.in/d2vTiwKP

The five minutes before a meeting are when you most want your network, and least have time to search it. Connect the Mesh MCP to Claude and you can ask for the whole briefing in one line: "Who am I meeting tomorrow, and what do I know about them?" What your assistant actually does with that: pulls tomorrow's events, finds each attendee in your network, and returns what you already have on them — where they work, how you met, your recent email history, and any notes you've saved. None of that is new information. It's all been in Mesh the whole time. The difference is that you no longer have to know which question to ask, or open four tabs to ask it. Our MCP server works with Claude, ChatGPT, Cursor, and Raycast: https://lnkd.in/ded2yg3U

A 2014 study asked students to track every conversation they had with classmates. Most days, they talked to about 2 people. The days they talked to more? Those were the days they felt happier and more connected. Casual acquaintances drove this effect more than deep friendships. The brief coffee. The hallway conversation. The person you don't know well but happened to run into. More interactions, even light ones, predicted better mood and stronger sense of connection. We tend to focus on going deeper with the people closest to us. This study suggests breadth matters just as much. The number of people you interact with on a given day. (You can read the full study at https://lnkd.in/gPRGBztV) The problem is we don't naturally track this. We remember the big conversations. The important meetings. But we lose track of the wider network—the people we used to see regularly who quietly drifted away. Mesh helps you see the full picture. Who you've been in touch with. Who you haven't. The people worth reaching back out to, not just the ones already on your calendar. More interactions with more people. That's where connection lives. Get started at https://me.sh

Being a good friend takes effort. Being disciplined about it, generous, present, the person who remembers—that doesn't come naturally to most of us. The pace of life gets in the way. Years ago we wrote down the method Mesh is built around. It's four things: ❶ Remember the important things. The kid's name, the book they mentioned, the thing they're worried about. Small details, but they're the raw material of feeling cared for. ❷ Embrace serendipity. The best moments happen by chance—running into an old friend abroad, meeting someone last-minute. Mesh is designed to manufacture a little of that on purpose. ❸ Do your research. The best conversations happen when someone's prepared. A one-minute review before a coffee changes the whole thing. ❹ Show up. The more you show up for people, the more good tends to find its way back to you. It's a practice, and it gets easier. None of this is complicated. But done consistently, it's most of what a lasting relationship is. The full method: https://lnkd.in/d2vTiwKP

Connect a few accounts to Mesh and you can end up with tens of thousands of contacts—every person you've ever emailed, followed, or saved a number for. Some people love having everyone in one place. Others want a tighter, more intentional network. If you're in the second camp, here's how to prune it. Archive the people you've drifted from. Search by recency—for example, everyone you haven't seen since before 2020—then select the whole list and archive it in one pass. Years of dormant contacts, gone in a few seconds. Let auto-merge handle the duplicates. In Settings, you can turn on Mesh's duplicate resolver, which automatically merges profiles only when it's highly confident they're the same person—matching name, same source, matching details. Anything it's unsure about, it leaves for you to review. It won't guess. Drop a whole integration if it's noise. If your Twitter import dragged in everyone you've ever followed, or iOS Contacts brought in a decade of one-off numbers, you can remove the entire source and keep Mesh focused on the people who actually matter to you. A smaller, truer network is often a more useful one. The full guide: https://lnkd.in/dAvGWC2A

Let's say you're traveling next week. You search your contacts for "New York". What you're really asking is: 1. "Who currently lives in NYC?" 2. "Who should I reach out to while I'm there?" Neither LinkedIn, nor your contacts, can answer those questions correctly. Mesh can. Most tools stop at the search result. They hand you 52 names that match your filter, then leave you to remember who each person is, why you connected, and whether it makes sense to reconnect. The same search means something different depending on what matters to you: • Planning a visit → who you'd genuinely enjoy seeing • Exploring a move → who could help you feel at home • Just passing through → who you've been meaning to catch up with Context changes everything. Mesh looks at your history with each person. Why you connected to begin with. When you last spoke. What you talked about that mattered. It surfaces the people it makes sense to reach out to, and why they came to mind—something LinkedIn or your Contacts app can't do. And Mesh works through AI tools like ChatGPT, Claude, Openclaw, and Hermes via our MCP and mesh-cli, making it the perfect home for your people and relationships. Ask "who should I see in NYC?" and get back context, not a filtered list. The people who matter, when it matters. Get started at me.sh today.

Mesh is even more powerful with the right strategies and weekly habits behind it. That's why we're excited to partner once again with the team at Beyond Connection, an immersive training that uses Mesh at its core. It kicks off next week! And who better to learn from than Keith Ferrazzi—who’s spent 20 years studying relationships and networks and written four bestsellers, including the NYT #1 bestseller Never Eat Alone. You'll learn alongside a curated group of professionals from 40+ countries, watching how each of them builds their own system around their own life — deals, career moves, fundraising, promotions, and more. There are only a handful of spots left, and this is the final week to apply. You'll fill out a quick application and schedule a call with their team to make sure it's the right fit for you — and mention you're a Mesh user for a tuition discount. Learn more and get started at https://lnkd.in/dSqTXgWX

One of the most valuable tech tools I use in my business led me to the best meal I had this weekend. It's my personal CRM and most recommended tool, Mesh. I used it to find Boston-based connections ahead of my recent trip to ask where I could find the best lobster roll in the city. Here's how I did it: • searched "location:Boston" • pushed down any feelings of awkwardness • sent LinkedIn messages to a few folks asking for their recommendation So so easy and not only did I get great food recs but I also got: • a great excuse to follow up with the folks who recommended spots I tried • a lunch date and several in-person convos with Paige Goldstein 🌀 • a list of recommendations to share with others • an opportunity to reconnect with people I haven't talked to in a while Checkout Mesh >> https://lnkd.in/gRyKF-8w

For the power users, tinkerers, and automators: there are now two ways to bring Mesh to the AI agents that increasingly work on your behalf — `clay-cli` in your terminal, and our Model Context Protocol (MCP) server in the assistant you already use. Clay in your terminal `clay-cli` is a command-line tool that lets you search, manage, and organize your contacts right from the terminal. It's fast, scriptable, and a natural fit for agent workflows — a small tool with a lot of possibility. Grab it on GitHub: https://lnkd.in/dEF7N6uP Mesh in your assistant MCP is an open standard designed to connect AI systems to various data sources and tools. Our MCP server lets assistants like Claude, ChatGPT, Cursor, and Raycast work directly with your network — searching contacts, retrieving past interactions, adding notes, and managing groups—all while keeping full control of your data. With the Mesh MCP, you can ask your agent things like: • "Who in my network works at Google as a Product Manager?" • "Who have I emailed the most this year?" • "Save a note for John Smith: 'Follow up about the Q2 report.'" • "Create a group called 'Investors 2025' and add Jane Doe." • "What meetings do I have scheduled for tomorrow?" To get started, enter our server URL https://mcp.me.sh/mcp into your client and follow the guided login flow — or connect through Smithery with the client of your choice. Documentation and a demo video are on GitHub: https://lnkd.in/ded2yg3U