Lead generation for startups: what to do before you can afford a sales team

Sunny Ray · Aug 20, 2026 · 7 min read

The best lead generation for an early startup is founder-led outbound to a tightly defined list: pick the exact companies and titles that buy from you, verify their emails, send one short message with one clear offer, and follow up on every reply the same day. Add content from your real conversations as it builds up. Do not hire a salesperson or buy expensive tools until you know which message turns strangers into conversations.

"Do things that don't scale" is the usual startup advice, and it is right as far as it goes. What it leaves out is what to do on Monday morning. Here is what we do, starting with the call on August 10 where Charles set the rule for the quarter.

The startup constraint: no team, no time

A startup founder doing lead generation has two problems. There is no one to delegate to, and every hour spent on outreach is an hour not spent on product or customers. So lead generation has to be narrow and efficient, not broad and busy.

That rules out a lot. Big content programs, events, paid ads with long tuning cycles, a dozen channels at once. It leaves a small set of things that produce conversations quickly: a good list, a good message and fast follow-up.

Start with a list you can name

On August 10 Charles mapped about 2,000 YC founders into my sending account. The list came from a database Amir, our teammate, had built a year earlier. Nobody had to scrape anything that week. It was already there, already structured, waiting to be used.

The rule Charles set alongside it was simple: strict ICP focus. Every conversation should hit the target. Not "anyone who might be interested", but people who match the profile of someone who buys.

That is the first thing I would tell any startup. Build a list you can describe in one sentence. "Founders of venture-backed B2B startups from the last three batches." "Heads of operations at logistics companies with 50 to 500 people." If you cannot name it that precisely, you will write a vague message, and vague messages get ignored.

Assets like Amir's database are worth building early. A list you own and keep clean is reusable for years: for customers, for guests, for investors.

One message, one offer

Your first message should make one offer that is easy to say yes to. For us, that offer is an invitation onto a show. For you, it might be a teardown, a pilot, an introduction, or a short conversation about a problem you know they have.

Rules that hold for any offer:

• About them, not you. The first line is about their company or their work.
• Short. Five sentences is plenty.
• One ask, one question, no links.
• From a separate sending domain, so your main domain stays clean.

Volume as the antidote to desperation

Here is something nobody says about early-stage selling: low volume makes you desperate. When you only have three conversations a week, every one of them feels like it has to close. You push. People feel it. They back away.

When you have thirty conversations, you can be honest. You can tell someone they are not a fit. You can walk away from a bad deal. You can ask better questions because you are not afraid of the answers.

That same week my own target was to close three deals while running pipeline from that ICP-strict list. That is only possible with enough qualified conversations that three closes is a normal outcome, not a miracle.

What to measure in month one

Keep it to five numbers:

1. People contacted.
2. Interested replies.
3. Conversations held.
4. Conversations with a dated next step.
5. Deals closed.

Ignore open rates. Ignore impressions. If conversations are low, fix the list or the message. If conversations are high but next steps are low, fix how you run the call. If next steps are high but closes are low, look at the offer and the price.

Should a startup outsource lead generation?

It depends on what you outsource. Outsourcing the plumbing makes sense early: list building, verification, sending infrastructure, inbox management. Outsourcing the conversations rarely does. Early buyers want to talk to the founder, and the founder needs to hear what buyers say.

The middle path is software that does the plumbing and drafts the replies while you stay in the conversations. That is the model we built.

How much should a startup spend on lead generation?

Less than a salesperson, until you know what works. Most early startups can test outbound for a fraction of one hire's cost. Spend on a clean list and good sending setup before you spend on headcount.

When to hire your first salesperson

Later than you think. Hire when you, the founder, have closed enough deals to know the message, the buyer and the objections, and you can hand someone a playbook rather than a hope.

Hiring before that is expensive in time as well as money. The Bridge Group's 2025 research puts average SDR ramp at 3.0 months and average tenure at 1.9 years. If you hire before you know the motion, a good chunk of that tenure goes to figuring out what you should have figured out yourself.

A useful test: could you write down the first message, the five most common objections and your answer to each, and the profile of the last ten customers? If yes, you are ready to hire. If not, keep selling yourself.

A simple first quarter

• Month 1: build the list, set up sending, start conversations, learn the objections.
• Month 2: tighten the ICP based on who replied and who bought. Double down on the message that works.
• Month 3: add one compounding channel, such as content from your conversations, and decide whether you are ready to hire.

That is not glamorous. It works because it keeps the founder close to the buyer, which is the one advantage a startup has over everyone else.

The mistakes I see most

The first is waiting to feel ready. Founders polish the website, the deck and the logo before they talk to a single buyer. Buyers do not care about the logo. They care whether you understand their problem, and the only way to learn that is to talk to them.

The second is an offer that asks too much. "Book a demo" is a big ask from a stranger. An invitation, a teardown or a short conversation about their problem is a small one. Start small and earn the bigger ask.

The third is stopping too early. Outbound rarely works in the first week. Sending domains need to warm, messages need a few rounds of testing, and replies trickle in. Give a channel a fair run before you decide it does not work for you.

The fourth is treating every lead the same. A founder who replies "interesting, tell me more" and a founder who replies "we need this next month" deserve different speed. Sort replies daily and answer the urgent ones first.

If you want to start without spending anything, connect Gmail and start free: about 10 invites a day written for you, saved as drafts you send yourself.

Frequently asked questions