< TRACTION

ambient (textambient.com) · traction 07

how to network as a founder: meet people, then go back to them

you've been told to meet more people. the best experiment we found suggests you should also go back to them.

7 min read4 nov 2026

start here

You started a company a few months ago. Someone has told you to network more, probably an investor or a friend who means well. You have a product that isn't finished, a first customer or two to keep, and maybe one free evening a week. A room of strangers with name badges is not where you want to spend it.

The advice you usually hear is to meet more people. It rarely says what to do after you've met them. That second part may matter more than you think. So don't only ask who you should meet next. Ask who you should see again.

what founders get from other people

Founders network for money, for customers and partners, and for advice from someone who has been through it. In a survey of venture investors, more than half of closed deals came through the investors' networks, almost 30% were deals they found themselves, and only about 10% came from the startup contacting the investor directly.

The harder question is how to spend your time on this, when you have so little of it.

what happened when owners saw the same people every month

In 2013, two economists, Jing Cai and Adam Szeidl, invited young firms in Nanchang, China, to join business groups. They split about 2,800 firms at random. The owners of some firms went into groups of 10 that met every month for a year, for about half a day each time. They toured one member's firm, then spent hours talking about borrowing, suppliers, hiring and marketing. The other firms had no meetings.

The members did no business with each other at the start, so these were new contacts, seen again and again. In the year of the meetings, their firms earned 8.1% more revenue than the others. Profit and the number of business partners rose too, and the effects were still there a year after the meetings ended.

Then the economists compared seeing people once with seeing them every month. They also put 466 of the same owners, at random, into extra groups of about 10 that met only once. After the year of meetings, the average owner had 2.18 more people recommending suppliers, clients, partners or staff from the monthly group than from the one-time group, and 1.16 more business partners. In the following year, the gaps were 2.4 people and 1.28 partners.

The monthly groups were meant to meet 12 times, so this wasn't an even contest. By our arithmetic, the one-time group brought more partners per meeting, so a single meeting isn't wasted. What repeat meetings added most was people willing to recommend others to you, and a recommendation is how you reach someone new. So going back and meeting new people aren't opposites.

two other experiments point the same way

In India, 100 founders of software startups were paired at random at a retreat, and each pair advised each other for about two days. Two years later, founders whose partner held regular meetings, set goals and gave frequent feedback had 28% more employees than founders whose partner did these things less often, and were 10 percentage points less likely to have failed. The effect was about twice as strong when the two founders were in the same city, and the authors think later advice may be part of the reason.

In Togo, business owners at a training program were paired at random for one conversation of 30 to 45 minutes with each of three others. Those paired with someone within a kilometer were 28 percentage points more likely to stay in touch over the next year. Owners who got to know three people whose businesses were, on average, about a kilometer closer earned about 10% more monthly profit. The authors think the gain came from the talks that followed, not only the first one.

the room with the name badges

We found no study of what a room of strangers with name badges does for a founder's money, customers or hires. The closest put electronic tags on 97 guests at one 80-minute mixer. Beforehand, 95% said they wanted new contacts. They then tended to stick with people they already knew, yet the average guest still met about seven strangers. Guests with the fewest friends there met the most strangers, so go alone.

Long afterwards, the authors asked two of the guests what had come of it. Both had mostly had no further contact with the strangers they met, though one had become good friends with one of them. An event can give you seven short conversations. Whether it was worth your evening depends on what you do next.

if you're shy

Research on shy founders is hard to find, so we looked at the nearest thing: people who had to network to find a job. In an experiment with 491 unemployed people, an online course of about 76 minutes raised confidence at networking, most of all for people who are less outgoing. People with more of that confidence got better jobs and pay, though they were no more likely to find a job. Once confidence was taken into account, how much networking they did made no clear difference to their results.

Talking to someone a second time is easier than walking into a room of strangers. You can start with a message, and you already know something about them.

if you have no time

The meetings in China took about half a day each, so about six days over the year. That sounds like a lot when you're building a company. But Cai and Szeidl estimate that the extra profit was twice the cost of organizing and attending.

If that is too much right now, start with one person. A message to someone you've already met takes a few minutes.

if you just started and know nobody

If you just started, you may feel you have nobody to go back to. But going back doesn't have to mean someone you met last month. It can mean a former colleague or classmate you haven't spoken to in years. When 224 executives were asked to contact two people they hadn't spoken to in at least three years, the 129 who finished said the help compared well with help from their current contacts.

If there is a specific person you need to reach, "How to ask for a warm intro to someone you don't know" walks through who to ask and what to send.

Another way to start is ambient, a private, text-native network where verified founders help each other solve problems and open doors. You text ambient, an AI agent, and say what you need. It sends your request to a founder who has already solved that problem. That founder does the work. You do the same for another founder, on something you know well. If the work goes well and you both say yes, ambient exchanges your names and numbers, so you can meet in person. So the first person you go back to is someone who has already done real work for you.

How do you get in? You can't buy your way into ambient, and you can't pitch your way in. A founder who's already inside invites you, or you text ambient and get verified. Before ambient looks for someone to help you, you help another founder or invite a founder who gets verified. That's how strangers become each other's unfair advantage. Tap "> access" at textambient.com.

what to do this month

Start with a list. Go through your calendar, email and phone, and write down everyone you talked to about your company in the last three months. Mark each person you have only spoken to once.

Pick three people on that list who told you something useful, or who have already done what you're about to do. Write to one of them today. Don't send "let's catch up", because that gives them nothing to answer. Tell them what you did with what they said, and ask one specific question.

Hi [first name], it's [your name] from [your company]. We met at [place] in [month], and you told me [what they said, for example to call the customers who left]. I tried it, and [what happened, for example two of the five told me why]. Now I'm stuck on [one problem], and I think you've seen it before. Could we talk for 20 minutes next week?

When you talk, ask what they actually did in their own company and what came of it. That tells you whether to take their advice. In India, what mattered was how often the partner set goals and gave feedback in their own firm. Before you finish, ask who else you should talk to about this. That is how one person you know leads to someone new.

Afterwards, do one thing they suggested and tell them what happened. Then you have a real reason to write again.

From now on, when you meet someone at an event, write down one thing they said that same day, and message them that week. Put 30 minutes in your calendar every Friday for these messages, so that following up doesn't depend on remembering.

If a few of these conversations go well, make them regular. Invite three or four founders who don't compete with you to meet once a month for a morning. Take turns hosting and show each other around, as the groups in China did. Each person brings one real problem, and at the next meeting says what happened. Someone else formed the groups in China, so expect to do the organizing yourself. Keep meeting new people alongside it.

At the end of the month, count again: how many people did you talk to for a second time?

sources

Paul Gompers, Will Gornall, Steven Kaplan and Ilya Strebulaev, "How Do Venture Capitalists Make Decisions?", Journal of Financial Economics, 2020. Working paper version. The deal shares come from 446 firm-level answers to that question. nber.org/system/files/working_papers/w22587/w22587.pdf

Jing Cai and Adam Szeidl, "Interfirm Relationships and Business Performance", Quarterly Journal of Economics, 2018. adamszeidl.com/papers/interfirm.pdf

Aaron Chatterji, Solène Delecourt, Sharique Hasan and Rembrand Koning, "When Does Advice Impact Startup Performance?", Strategic Management Journal, 2019. 2018 author version. hbs.edu/ris/Publication%20Files/advice-impact-startup_b2f0105d-95d3-4f6d-8ae1-63f54c868548.pdf

Stefan Dimitriadis and Rembrand Koning, "Networking Frictions and Entrepreneurial Learning in Developing Economies", Management Science, published online 10 July 2024. stefandimitriadis.com/_files/ugd/41d61d_8a07541711044e8495442944deaead42.pdf

Paul Ingram and Michael Morris, "Do People Mix at Mixers?", Administrative Science Quarterly, 2007. Authors' draft. business.columbia.edu/sites/default/files-efs/pubfiles/2552/pingram.pdf

Connie Wanberg, Edwin van Hooft, Songqi Liu and Borbala Csillag, "Can job seekers achieve more through networking?", Personnel Psychology, 2020. carlsonschool-d8.prd.umn.edu/sites/carlsonschool.umn.edu/files/2020-12/Wanberg%20Van%20Hooft%20Liu%20Csillag%202020%20PPsych.pdf

Daniel Levin, Jorge Walter and J. Keith Murnighan, "Dormant Ties: The Value of Reconnecting", Organization Science, 2011. business.gwu.edu/sites/g/files/zaxdzs5326/files/15_FP.SP_Walter.J_15levin_2011a.pdf

More articles: traction. More questions: FAQ.