You've got an idea. Maybe a prototype. Maybe a few early conversations with potential users. But you don't have a marketing team, a brand, a content library, a social media following, or a budget. You're pre-seed, pre-everything, and someone told you that you need a "go-to-market strategy." The advice online is either too generic ("build your brand!") or too advanced ("optimize your CAC/LTV ratio!"). Neither helps when you're starting from literal zero.

I've been here — not as a founder, but embedded with founders at this exact stage. When I started working with Drebbel, they were two technical co-founders with a working prototype and zero marketing infrastructure. No website beyond a one-page landing page. No social accounts. No email list. No content. The question wasn't "how do we scale?" It was "where do we even begin?"

This post is the plan I wish existed when I started doing this work. It's the sequence of steps I now use with every pre-seed AI startup I work with. Not a framework — a step-by-step plan you can execute starting today.

Before You Do Anything: The Three Questions

Before you write a single post or send a single email, you need clear answers to three questions. Everything else builds on these.

  1. Who is your specific user? Not "product managers" or "small businesses." A specific person at a specific type of company with a specific problem. Example: "Head of customer support at B2B SaaS companies with 20-100 employees who spends 3+ hours per day manually categorizing support tickets."
  2. What changes for them when they use your product? Not features. Outcomes. "They go from spending 3 hours categorizing tickets to spending 15 minutes reviewing AI-categorized tickets, and their response time drops by 40%."
  3. Where do they already spend time online? Which communities, platforms, publications, and events are they part of? You need to know at least five specific places.

If you can't answer these precisely, that's your first task. Talk to 10-15 potential users. Not to pitch them — to understand their world. Where they hang out, what frustrates them, how they currently solve (or fail to solve) the problem you're addressing. These conversations are the foundation of everything that follows.

I've written a comprehensive go-to-market strategy guide that covers the strategic thinking behind these decisions. But this post is about execution — what you actually do, week by week.

Month 1: Foundation (Weeks 1-4)

The first month isn't about getting users. It's about building the minimum infrastructure you need to start getting users in month two. I know that sounds slow, but skipping this step is why most pre-seed startups waste their first three months flailing.

Week 1: Positioning and Messaging

Write three things:

  • A one-sentence pitch: "[Product] helps [specific user] do [specific outcome] by [how it works]." Keep it under 15 words if you can. This is what you'll use in every cold email, every LinkedIn bio, every community introduction.
  • A one-paragraph description: Expand the pitch into 3-4 sentences that cover the problem, your solution, and why it's different. This goes on your landing page.
  • Three proof points: Any evidence that your product works. Early user quotes, demo results, benchmarks against manual processes, anything. If you have nothing yet, that's okay — your first task is getting proof points through design partner conversations.

Don't overthink this. Your positioning will change as you learn more. But you need a starting point, and "AI-powered platform that leverages cutting-edge technology" is not a starting point. Be specific, be concrete, be boring if necessary. Clarity beats cleverness every time at this stage.

Week 2: Minimum Viable Online Presence

You need exactly four things:

  • A landing page with your one-paragraph description, a screenshot or demo GIF, and a signup form. That's it. No pricing page, no blog, no about page. One page.
  • A LinkedIn profile (for the founder) that clearly states what you're building and for whom. Update your headline, about section, and featured content.
  • An email address at your domain (not Gmail). It doesn't matter that your company is two people. firstname@company.com signals legitimacy in a way that company.ai@gmail.com never will.
  • A simple way to track signups. A spreadsheet is fine. You need to know who signed up, when, and where they came from.

That's your entire marketing infrastructure for month one. Resist the urge to build more. You don't need a blog yet. You don't need social media accounts for the company. You don't need a newsletter. All of that comes later, once you know what to say and who to say it to.

Weeks 3-4: First Outreach

Now you start talking to people. Not marketing to them — talking to them. The goal for weeks 3-4 is to have 20-30 genuine conversations with potential users. Here's how:

  • Personal network: Message everyone you know who might be relevant. Not a mass email — individual, personal messages. "Hey, I'm building something for [type of person]. Do you know anyone who deals with [specific problem]?"
  • LinkedIn outreach: Send 10 connection requests per day to people who match your ideal user profile. When they accept, start a conversation about the problem space. Not a pitch — a conversation.
  • Community lurking: Join the five communities you identified earlier. Don't post yet. Spend two weeks reading, understanding the culture, and identifying the people and threads that are relevant.

By the end of month one, you should have a clear landing page, an active LinkedIn presence, 20-30 conversations with potential users, and (hopefully) your first 10-20 signups from personal outreach.

Month 2: First Traction (Weeks 5-8)

Month two is when you start to actively market with no budget. The foundation is set. Now you build on it.

Week 5: Launch Your Experiment Engine

Set up a simple growth experiment tracker. A spreadsheet with these columns: experiment name, channel, hypothesis, start date, end date, metric, result. Every week, you'll run two to three experiments and log the results.

Your first round of experiments should test the channels where your users hang out. Based on what I've seen work for AI startups at this stage, start with these three:

  1. Community contribution: Start posting in the communities you've been lurking in. Share genuinely helpful content about the problem you solve. Don't mention your product yet.
  2. Cold outreach: Start sending 10-15 personalized emails per day to potential users. Use the one-sentence pitch you wrote in week 1.
  3. LinkedIn content: Post twice per week about the problem space. Share insights from your user conversations (anonymized). Be a person with opinions, not a company with announcements.

Weeks 6-7: Double Down and Cut

After two weeks of experiments, look at the data. Which channel generated the most signups? Which had the best conversion rate? Which felt most sustainable?

Cut the worst performer and redirect that time to the best performer. This is the most important discipline in pre-seed marketing: saying no to things that kind-of work so you can focus on things that really work.

At Drebbel, we tested five channels in month two. Three produced nothing. One produced a trickle. One — cold outreach to heads of engineering at mid-size companies — produced a flood. We dropped everything else and did nothing but cold outreach for the next three weeks. That's how we went from 15 users to over 100.

Week 8: Tell Your Story

By week 8, you have something most pre-seed startups don't: data and stories. You know which channels work. You've had real conversations with real users. Some of them are getting real value from your product.

Now it's time to start telling that story publicly. Write your first piece of substantial content — not a company blog post, but a genuine insight piece based on what you've learned. Post it on LinkedIn. Adapt it for relevant communities. Send it to the contacts you've been building relationships with.

The content should follow a simple formula: "We talked to X people dealing with Y problem, and here's what we learned." This is incredibly effective because it demonstrates expertise without being salesy. You're sharing learnings, not pitching a product. The product is implied.

Month 3: System Building (Weeks 9-12)

By month three, you should have 50-200 users (depending on your market), a clear understanding of which channels work, and a library of conversations and insights. Now you turn the scrappy hustle into a repeatable system.

Build Your Content Engine

Now — and only now — it's time to start thinking about content as a system. Based on what you've learned from user conversations and community engagement, you know exactly what topics resonate. Start creating content around those topics.

The pre-seed content strategy is simple: one substantial piece per week, distributed across three to four channels. I've written a full guide on AI startup marketing that covers the content side in depth, but the key principle is this — create content that answers the specific questions your target users are already asking. Don't guess. Use the conversations you've had to know exactly what those questions are.

Systematize Your Outreach

If cold outreach is working (and for most B2B AI startups, it will be), it's time to build a system around it. That means:

  • A repeatable prospecting process for finding new targets weekly
  • Email templates (not scripts — templates with room for personalization)
  • A follow-up cadence that runs semi-automatically
  • A tracking system that connects outreach to signups to activation

You don't need expensive tools for this. A spreadsheet and a Gmail account will get you to 200+ users. The tools come later, when you're spending more time on admin than on actual outreach.

Start Building Your Path to 1,000 Users

By the end of month three, you should have a clear growth model. You know your cost per acquisition (in time, since you have no budget). You know your activation rate. You know which channels scale and which don't. You can project a rough timeline to 1,000 users.

If that timeline is longer than six months, something needs to change. Either your channels aren't efficient enough (experiment more), your activation rate is too low (fix the product experience), or your market is harder to reach than you thought (reconsider positioning).

What Not to Do at Pre-Seed

I want to be equally clear about what to avoid, because I see pre-seed founders waste critical time on things that don't matter yet:

  • Don't hire a marketing agency. They'll spend your money on generic tactics that work for established companies and fail for pre-seed startups. You need founder-led marketing at this stage.
  • Don't build a brand. Your brand is your product and your founder's reputation. Logos, brand guidelines, and visual identity can wait until you have product-market fit.
  • Don't invest in SEO. It takes 6-12 months to pay off. You need signups now, not in a year. SEO comes after you've validated the product and the market.
  • Don't run paid ads. You don't know your positioning well enough yet to write ads that convert. Use organic channels to test messaging first, then consider ads post-seed.
  • Don't build an elaborate tech stack. You don't need a CRM, a marketing automation platform, an analytics suite, and a social media scheduler. You need a spreadsheet and the willingness to do things manually.

The Pre-Seed Marketing Mindset

The most important thing I can tell you about pre-seed marketing isn't a tactic or a channel. It's a mindset shift. At pre-seed, marketing is not about scaling. It's about learning. Every email you send, every community post you write, every conversation you have — the primary purpose is to learn what resonates, who responds, and why.

The startups that move fastest through the pre-seed phase are the ones that treat marketing as a learning engine. They run small experiments, measure everything, talk to users obsessively, and adapt their approach weekly based on what they find.

The ones that struggle are the ones trying to "do marketing" the way they've seen established companies do it. They spend weeks on a beautiful website nobody visits. They write blog posts optimized for keywords nobody's searching. They build social media accounts that get 12 likes per post from other founders.

Pre-seed marketing is scrappy, manual, and relentlessly focused on direct user conversations. It doesn't look like marketing from the outside. It looks like talking to people. And that's exactly what it should be.

If you're past this stage and looking for the bigger strategic picture, read my full AI startup marketing guide. And if you're ready to think about your go-to-market strategy more formally, I've written a founder's playbook for that too. But if you're truly at the beginning — pre-seed, pre-traction, pre-everything — start here. Do month one. Then do month two. Then month three. One step at a time, one experiment at a time. That's how every successful startup I've worked with got off the ground.