Koka Sexton

Last Tuesday morning I sat down at my desk, opened my outbound queue, and wrote 27 personalized outreach drafts in one session. Not a single one started with “Hi {first_name}.” Not a single one was templated. And not a single one required me to type a recipient’s name into a compose window.

I didn’t do this by working faster. I built a system.

TL;DR

• Outbound is broken because most teams treat personalization as a mail-merge variable. Real personalization comes from audience intelligence.
• I built a 6-stage lead flow system that takes prospects from selection to sent draft without manual data entry.
• The dual-offer model splits outreach by company vs. personal audience size, matching the right offer to the right signal.
• Every interaction feeds back into a 14-factor ICP scoring model, so your CRM gets smarter with every send.
• You don’t need more leads. You need the right system for the ones you already have.

---

The Problem With Outbound Is Structural, Not Tactical

Most outbound sequences are dead on arrival. The average cold email gets a sub-2% response rate. The average LinkedIn InMail fares worse. And the average sales team responds by doing more of the same — more sends, more sequences, more automation layered on top of a broken premise.

The broken premise is this: that personalization means inserting a first name into a template.

I’ve spent the last two years testing every outbound tactic available. Sequences from HubSpot. Cadences from Outreach. AI-generated intros from every vendor that raised a Series A between 2021 and 2024. None of them solved the fundamental problem, which is that people know when they’re being mail-merged. Your prospect can tell the difference between someone who researched their business and someone who copy-pasted their name into field 3 of a 12-step sequence.

The fix isn’t a better template. It’s a system that gives you real data to personalize from — without requiring you to spend 15 minutes per prospect doing manual research.

People know when they’re being mail-merged. Your prospect can tell the difference between someone who researched their business and someone who copy-pasted their name into field 3 of a 12-step sequence.

The 6-Stage Lead Flow System

Here’s the architecture I built. It processes prospects through six stages, each one feeding into the next.

---

Personalization That Actually Works

Let me show you what real personalization looks like in practice.

When my system processes a prospect, it doesn’t insert a variable. It gathers actual profile intelligence:

• Follower counts and follower-to-following ratios (signal of audience relevance)
• About section content (what they say about themselves, not what a database says)
• Professional headline (current role, stated priorities)
• Recent post activity (are they active? what are they talking about?)
• Mutual connections (context for the opening line)

That data doesn’t get stuffed into brackets. It shapes the entire message. The difference between “I see you’re the VP of Sales at Acme Corp” and “Your LinkedIn audience has grown 40% in six months — curious how you’re thinking about converting that attention to pipeline” is the difference between delete and reply.

The Dual-Offer Model: Matching the Right Message to the Right Signal

One of the biggest mistakes in outbound is sending the same offer to everyone. A CMO with a 50,000-follower company page has different needs than a founder with a 2,000-follower personal profile. The system should recognize that and route accordingly.

The routing logic is straightforward: if the audited company page has more than 15,000 followers, the outreach opens with a SignalScout angle. If the individual prospect has a personal following above 12,000, the VCO framework leads. If both apply, the system prioritizes the stronger signal.

This isn’t a split test. It’s signal-based routing. The offer matches the data point that matters most to the recipient.

---

The Email Style: Hook → Insight → CTA

Every outbound message follows the same architecture, regardless of which offer it carries:

Notice what’s missing: no em dashes, no exclamation points, no faux enthusiasm. Just data, framed cleanly, with an open door at the end.

Infrastructure That Doesn’t Break at Scale

Most outbound workflows collapse when volume increases. The system I built doesn’t have that problem, because it was designed for scale from day one.

Domain-wide email. Seven verified aliases on the kokasexton.com domain. Each one can send and receive independently. This isn’t about hiding — it’s about resilience. One alias hits a spam filter? The rest keep running. Different contexts get different senders. Press communications go from press@. Sales outreach from sales@. Personal notes from koka@.

Sent-mail monitoring. Every outbound message is logged and timestamped automatically. No “did I already email this person?” No duplicate sends. No wondering what stage a conversation is in.

100% follow-up rule. If a prospect hasn’t been touched in 14 days, they surface. Not based on a sequence timer. Based on actual engagement signals. The system doesn’t let anyone fall through the cracks because someone forgot to check a dashboard.

---

The Compounding Effect: Why the CRM Gets Smarter Every Send

Here’s the part most people miss: the system doesn’t just send emails. It learns.

Every interaction — every draft, every send, every reply, every profile view — feeds back into a 14-factor ICP scoring model. Did they open? Did they reply? Did they view your profile within 48 hours of receiving the message? Did they visit a specific page on your site?

Each signal adjusts the contact’s score. Over time, the CRM builds a heat map of who’s warming up, who’s ready for a conversation, and who needs more time. You stop guessing. You start working from data.

This is the difference between a lead list and a living pipeline. A static list degrades the moment you export it. A scoring model improves the moment you use it.

Every interaction feeds back into a 14-factor ICP scoring model. A static list degrades the moment you export it. A scoring model improves the moment you use it.

What This Looks Like in Practice

Last Tuesday, I ran a full session. Here’s what happened:

• Audience audit identified 27 high-signal prospects from existing follower lists and demand generation signals
• Profile data was gathered for each — About sections, headlines, follower counts, recent activity
• 27 personalized drafts were generated, each referencing real data points from the prospect’s profile
• Each draft followed the Hook → Insight → CTA structure, routed to the appropriate offer (SignalScout or VCO)
• Drafts went to review, not straight to send — the human is always in the loop for final approval

Total time: one focused session. Not 27 separate research sessions. Not “I’ll do five today and five tomorrow.” One session, 27 drafts, zero template variables.

This is what signal-based prospecting looks like when you stop treating it as a buzzword and start treating it as an operating system.

---

You Don’t Need More Leads

The outbound industry has spent a decade optimizing for volume. More lists. More sequences. More sends. The result is inboxes full of garbage and response rates that make a lottery ticket look like a sound investment.

The teams winning outbound right now aren’t sending more. They’re sending smarter. They’re building systems that do the heavy lifting — the research, the personalization, the routing, the follow-up logic — so the human only touches what requires human judgment.

Twenty-seven personalized drafts in one session isn’t a productivity hack. It’s what happens when you stop treating outbound as a volume game and start treating it as an intelligence game.

You don’t need more leads. You need the right system for the ones you already have.