How to write a cold message that gets replies: 3 rules from 10+ YC startups

Most cold messages get ignored, and it's usually the same few reasons every time. I run GTM for 10+ YC startups in San Francisco, and the messages that actually get replies follow three simple rules. None of them are clever. They just respect the fact that a stranger owes you nothing and has about two seconds to decide whether to keep reading.

Rule 1: Open with the problem in one line

Start with the problem the other person might be having, said plainly. Not your product, not your company, the thing that's annoying them.

If you sell to founders building with AI, that opening line can be as simple as: "Are you trying to ship your next MVP but find all the tools confusing?" If they feel that, they keep reading. If they don't, they were never going to buy anyway, and you just saved both of you time.

The mistake is opening with who you are and what you built. Nobody has agreed to care about that yet. Lead with the problem and you've earned the next sentence.

Rule 2: Show why you're worth listening to

Right after the problem, give a quick reason you're the right person to be talking about it. One line of proof. "We've got 2,000 stars on GitHub." A relevant past job. Where you studied. A result you got for someone like them.

This is the part that separates a message worth answering from noise. Anyone can name a problem. The reason someone replies is that you sound like you actually understand it and might be able to help. If you have a real customer result, use it, the same before-and-after proof that works on a landing page works in a cold message.

Rule 3: Never ask for a 15-minute meeting

This is the biggest one. Asking for a meeting in your first message is the fastest way to get ignored. Think about it from their side: if a stranger walked up to you and asked for 15 minutes, you'd feel a little creeped out. A cold inbox is the same.

Ask a question about their problem instead. If your product helps companies cut their AI inference spend, ask: "Is AI spend something you're thinking about this quarter?" Now they can say yes or no, and either way you've started a conversation. A meeting ask gives them nothing to do except delete it.

The whole goal of the first message is a reply, not a calendar invite. Get the conversation going and the meeting comes on its own.

Bonus: stop over-personalizing

The number one mistake I see in AI-generated cold messages is fake personalization. Opening with something the person posted on LinkedIn, or their job title, or their career, everyone can tell you pasted it out of ChatGPT and never actually looked at their profile. It reads as a template, because it is one.

You have very little text before someone tunes out. Don't spend it proving you skimmed their LinkedIn. Spend it on the problem and on why you're the right person to solve it. That's what earns a reply. If you want a full checklist for cutting the tells out of a message, we wrote up how to review AI-written cold messages.

Put the rules together

A message that follows all three looks like this: one line naming the problem, one line of proof that you get it, and one question they can actually answer. No intro paragraph, no "hope you're well," no meeting ask. That's it.

The reason I'm confident about this is that it's what's working right now for the YC founders I help book meetings and close deals. Running these messages across LinkedIn, email and X on the same buyer, and changing them based on who replies, is exactly what we build PumpGTM to do. If you want the playbook top San Francisco startups are using, start at pumpgtm.com.

Transcript

Here's three simple components of a great cold message, from helping 10+ Y Combinator startups. Number one, start with the problem in one simple line. Talk about the problem the other person might be encountering. So if you've got a YC coding startup, something you can ask is: hey, are you trying to build your next MVP but find all the tools confusing? Number two, always include some evidence or authority, proof of why you're relevant to this problem. Something simple. You can say, hey, I've got 2,000 stars on GitHub. Or you can talk about your college, your past job, anything to give evidence of why you're the right person to talk about this. Number three, never ask for a 15-minute meeting at the end of it. That's one of the biggest mistakes I'm seeing everyone make. If you ask someone for a meeting, even in your first cold message, honestly, that's a little weird. Think about it this way: if a stranger came up to you and asked for a 15-minute meeting, wouldn't you feel a bit creeped out? So the better thing to do is ask them a question about the problem they're having. For example, if your startup helps companies save on AI inference spend, you can simply ask: hey, is AI spend something you're thinking about this quarter? Because if they say yes or no, you can at least start a conversation, versus asking for a 15-minute meeting and getting ignored. And a bonus tip: the number one mistake I'm seeing in these AI-generated cold messages is over-focusing on personalization. Talking about someone's job, their career, or something they posted on LinkedIn, everyone knows you just copied and pasted it from ChatGPT and didn't actually check their profile. Stop wasting crucial text. You have very little time to get their attention, so use it by talking about the problem and your authority in presenting the solution. The reason I'm so confident with this advice is that I'm here in San Francisco helping 10+ of the latest Y Combinator startups automate their GTM and actually get meetings and sales booked on the calendar. I'm Namanyay, the CEO of PumpGTM, and if you want the same playbook top startups in San Francisco are using, check the links in the description. Thank you.

Questions founders ask

How do you start a cold message? With the problem the person has, in one plain line, not with who you are or what you built. If the opening line makes them think "that's me," they keep reading.

Should I ask for a meeting in a cold email? No. A meeting ask in the first message is the most common reason cold outreach gets ignored. Ask a question about their problem instead, so they have something easy to answer and a conversation can start.

How much should I personalize a cold message? Less than you think. Referencing someone's LinkedIn post or job title usually reads as a template pasted out of ChatGPT. Spend your limited text on the problem and on why you're credible, not on proving you skimmed their profile.

What makes a prospect actually reply? A problem they recognize, a quick reason to trust you, and a question they can answer in one line. Real proof, like a customer result or a concrete number, does more than any amount of flattery.

Does this work for LinkedIn and email both? Yes. The three rules are about the message itself, so they hold across LinkedIn, email and X. What changes is length and tone, not the structure.