Let me tell you about a number that changed how I think about outreach. When I started working with Drebbel, their cold outreach response rate was sitting at 2%. Two out of every hundred people they contacted bothered to reply. Most of those replies were "not interested." Sound familiar?
Within a few months, we got that number to 30%. Not by writing better subject lines. Not by using some magic email tool. We got there by fundamentally rethinking who we were talking to and why they should care.
Most founders I talk to assume their outreach problem is a copywriting problem. It almost never is. It's a targeting problem. And fixing targeting changes everything downstream.
Why Most Outreach Fails (And It's Not Your Copy)
Here's the pattern I see constantly: a founder builds a great product, writes what they think is a compelling cold email, blasts it to a list of 5,000 contacts scraped from LinkedIn, gets a 1-3% response rate, and concludes that "cold outreach doesn't work for our product."
But look at what actually happened. They sent the same message to 5,000 people with wildly different needs, roles, and pain points. Some of those people had the exact problem the product solves. Most didn't. The message was fine. The audience was wrong.
Bad targeting is the silent killer of outreach campaigns. You can have the best copy in the world, but if you're sending it to someone who doesn't have the problem you solve, you're just generating spam. And spam doesn't convert.
The first thing I did at Drebbel wasn't to rewrite their emails. It was to throw away their contact list and start over.
Building an Ideal Customer Profile That Actually Works
Everyone says "know your ICP." Very few people actually do the work. An ideal customer profile isn't a persona document that says "VP of Engineering at a mid-size company." That's a job title, not a profile.
A real ICP answers these questions:
- What specific event or situation makes someone need your product right now? (Not in theory. Right now.)
- What are they currently doing to solve this problem? What's painful about that solution?
- What signals can you observe externally that indicate someone is in this situation? Job postings, funding rounds, tech stack changes, team growth, public complaints?
- Who in the organization feels this pain most acutely? Not who signs the check. Who loses sleep over the problem?
When we rebuilt the ICP at Drebbel, we went from "companies that might need our product" to "companies that just hired their third data engineer and are posting about data pipeline reliability on LinkedIn." That's a completely different level of specificity. And it transforms every message you send.
If you're in the early stages of defining your market, your go-to-market strategy and your ICP work should be happening at the same time. They feed each other.
LinkedIn Strategy: Playing the Long Game
LinkedIn is the most underrated outreach channel for B2B SaaS, but most people use it wrong. They send connection requests with a sales pitch in the first message. They spam InMails. They treat it like cold email with a different interface.
Here's what actually works, especially for AI founders:
The Warm-Up Sequence
Before you ever send a direct message, your target should have seen your name at least three times. This isn't stalking. It's basic marketing psychology. People respond to people they recognize.
- Week 1: Engage genuinely with their content. Leave thoughtful comments on their posts. Not "Great post!" but actual substantive responses that add to the conversation.
- Week 2: Share content they'd find relevant and tag them, or reference their work in your own posts. Show that you're in their world.
- Week 3: Send a connection request with a short, personal note referencing a specific thing they said or did. No pitch.
- Week 4: After they accept, send a message that opens with value, not an ask. Share a resource, an insight, or a data point relevant to their specific situation.
This takes longer than blasting connection requests. It also works about fifteen times better. When someone already recognizes your name, your message gets read. When you're a stranger, it goes straight to the filtered inbox.
Content as Outreach
The best LinkedIn strategy for AI founders is publishing content that demonstrates expertise in your prospect's problem space. Not product announcements. Not company updates. Genuine insights about the problems your target customers face.
When you consistently publish useful content, two things happen: inbound starts flowing because people discover you through the algorithm, and your outbound becomes dramatically more effective because prospects check your profile before responding, and what they find is evidence that you actually understand their world.
Email Sequences: Structure That Converts
Once you have the right targets and you've done the warm-up work, your email sequences need structure. Here's the framework I've used across multiple B2B SaaS campaigns:
Email 1: The Observation
Open with something specific you noticed about their company. A recent hire. A product launch. A public challenge they mentioned. Then connect that observation to the problem you solve. No pitch. Just, "I noticed X, and in my experience, companies in that situation often struggle with Y."
Email 2: The Case Study (3 days later)
Share a brief, specific result from a similar company. "We worked with [Company] when they were in a similar stage. They were spending 20 hours a week on [painful task]. We got that down to 2." Let the result speak for itself.
Email 3: The Resource (5 days later)
Send something genuinely useful with no strings attached. A guide, a checklist, a template. Something they can use whether or not they buy your product. This builds trust and demonstrates expertise.
Email 4: The Direct Ask (7 days later)
Now you can ask for the meeting. But make it easy to say yes. "Would 15 minutes this Thursday work to see if this is relevant for your team?" Low commitment, specific time, clear purpose.
The key is that by email 4, you've provided value three times. You're not a stranger asking for something. You're someone who's been helpful, and now you're suggesting a conversation.
Personalization at Scale: The Practical Approach
The biggest objection I hear is "but personalization doesn't scale." And if you mean writing every email from scratch, that's true. But there's a middle ground that works.
Layer your personalization:
- Segment level (set once per list): Industry-specific pain points, relevant metrics, common objections.
- Company level (1 minute per company): Recent funding, headcount growth, tech stack, public initiatives.
- Person level (30 seconds per person): Their role, a recent post or comment, a shared connection.
Build templates for each segment, then spend 90 seconds per contact adding company and person details. At that rate, you can personalize 30-40 emails per hour. That's plenty when your targeting is tight, because you don't need to contact thousands of people. When you're reaching the right 200, those 200 matter more than the wrong 5,000.
Getting your first 1,000 users often starts with exactly this kind of focused, personal outreach rather than broad-blast campaigns.
The Follow-Up System: Where Money Lives
Here's a stat that should change your behavior: 80% of sales require at least five follow-ups, but 44% of salespeople give up after one. If you're not following up systematically, you're leaving the majority of your potential responses on the table.
But following up doesn't mean sending "just bumping this to the top of your inbox" four times. Each follow-up should add something new:
- A new data point relevant to their situation
- A piece of content that addresses a likely objection
- A change in their company that creates new urgency
- Social proof from a competitor or peer who just signed up
Set up a simple CRM pipeline. Every contact gets tracked. Follow-ups are scheduled automatically. Nothing falls through the cracks. The system does the remembering so you can focus on the conversations.
Measuring What Matters
Most people track open rates. Open rates are vanity metrics, especially now that email privacy features make them unreliable. Here's what to actually measure:
- Response rate by segment: Which ICP segments reply most? Double down on those.
- Positive response rate: A reply of "not interested" isn't the same as "tell me more." Track them separately.
- Meeting booked rate: Of those who respond positively, how many actually get on a call?
- Pipeline generated per hour of outreach: This is the number that tells you whether outreach is a good use of your time compared to other channels.
At Drebbel, the 30% response rate was great, but the number that actually mattered was that 12% of all contacts became booked meetings, and the pipeline generated per hour of outreach was 4x higher than any other channel. That's the kind of data that tells you to invest more, not just feel good about a metric.
Making Outreach Part of a Bigger System
Cold outreach works best when it's not operating in isolation. It should be one piece of a broader go-to-market strategy that includes content, community, and product-led growth. The prospect who gets your cold email and then sees your LinkedIn content and then reads a blog post is exponentially more likely to convert than someone who only encounters you through a cold message.
If you're an AI startup founder trying to decide whether to invest in outreach versus other channels, the honest answer is you probably need both. Outreach gives you immediate conversations with specific people. Content and product-led growth build the long-term engine. The startups that grow fastest figure out how to make these channels reinforce each other.
Need help building this kind of integrated growth system? That's exactly what a fractional head of growth does: stitch together outreach, content, PLG, and launches into something that compounds.