Email sequences based on intent data

I built a system that uses an AI Agent and Claude to write hyper-personalized outbound messages for 1,000s of prospects based on their LinkedIn data and online activity. In this post, I'll show you exactly how this system works. (Watch the video to see the full system in action) You'll see how to write outbound messages that make it look like you spent 30 minutes researching each prospect before sending any message (when it's actually all automated). Here's the complete process: 1. Finding the right prospects We start by using Clay to build a list of perfect-fit prospects. Basic info like name, company, LinkedIn URL, work emails, etc., can be easily found, but we need more data for truly personalized outreach. 2. Using AI Agent to gather relevant data points Our AI Agent visits each prospect's LinkedIn profile and company website and also searches them on the internet to find relevant data points. This gives us at least 5 unique data points for each person, which we can use in our messaging. 3. Feeding everything to Claude Next, we take all those data points along with their LinkedIn data and feed them to Claude Anthropic using a specific prompt. The prompt is over 6000+ characters (pretty long), and the prompt's format includes sections for company data, personal info, and targeting guidelines that make each message unique. 4. Writing complete outbound sequences (Important) The system doesn't just write one email or message - it creates entire sequences, including initial cold emails, follow-ups, and LinkedIn messages. Each message references details about the prospect's role, work history, and company challenges. Because Claude has all their data points, every follow-up stays relevant to who they are and what they care about - not generic templates with just a name/company name swapped in. 5. How to set this up yourself The setup process is straightforward. You'll need: - A Clay account to build prospect lists - Access to an AI Agent tool for research (You can use Claygent for that) - Claude for writing the messages (You must know prompt engineering) - Clear understanding of who your ideal customers are _____ Once it is set up, the entire process can be automated. Then, you don't have to worry about spending hours manually researching prospects or writing generic emails or LinkedIn messages that get ignored.

Teams who take a “boil the ocean” approach to outbound will fail. Here’s how to fix it and build sequences that actually drive results: Step 1: Focus your team on accounts most likely to buy now, invest at a premium, and become long-term customers or referral sources. This means moving beyond “anyone who fits the ICP” and zeroing in on high-priority targets. Step 2: Create deeper, more meaningful segments from that refined group. Traditional segments are great for organizing territories but fall short for crafting sequences that resonate. Instead, you need segmentation that helps your team speak the language of specific sub-groups. Use multiple layers of data—firmographics, intent signals, and contact-level insights—to break your TAM into smaller, actionable groups. Step 3: Launch micro-campaigns that target those precise segments with messaging designed to feel tailor-made. When you take this approach, personalization becomes scalable because it’s rooted in segmentation. Your reps don’t waste time on one-off customization, and your messaging feels 99% relevant to the prospect. I've been teaching this process as #ValueBasedSegmentation for the better part of a decade. It’s the key to building sequences that drive higher CTRs, replies, and engagement without tedious manual effort. ➡️ With this approach, you’ll: - Improve email performance - Write copy that prospects actually care about - Give your team a clear roadmap for focused outbound 📌 How are you helping your team build relevance into their outbound sequences?

Month 1 of building our outbound system from scratch. Here's the game plan. I built a bunch of outbound systems (one took ARR from $1M→$14M in 2 years) But back then, the tech was clunky. No Clay, no AI, no smooth workflows. Now, the outbound game is the same. But we are playing with better gear. My goal - use it wisely. It’s easy to chase tools and never ship. Tool FOMO is a great procrastination hack. So instead, let's start with foundations: Rule #1: Lead list quality > everything Every lead must pass the 3Qs filter: Why now (evidence they’re buying), Why them (ICP + ACV match) and Why you (unique fit). I'm using multiple intent tools and Clay to score and prioritise accounts with real buying energy, not just a job title. Rule #2: Match ACV to effort High ACV → LinkedIn + calls + email + manual Mid/Low ACV → email-first at scale Outbound isn’t blasting; it’s matching effort to potential return. Now, volume vs quality? If you have huge TAM + lots of competitors + many low-ACV leads: Don’t ignore them while you romance your top 50; time passes, intent cools, a competitor wins. Tiny TAM: Do the deep work on every account - you don’t have the luxury to burn them. Rule #3: Structure lists for campaigns (not the other way around) Group by buying signal/source for relevance at scale, by ICP lane (Sales Teams vs Agencies) for tailored value props, by (warm vs cold) for easy personalization hooks, and by role (champion vs DM) for outcome-specific copy. Good list segmentation, easy campaign creation. Rule #4: Build the ladder: low-hanging → warm → cold Re-engage trials/churned with intent, then warm engagers (events, followers, visitors), then cold ICP with triggers. Sequence the work so you earn quick wins and learn fast. Rule #5: Triggers beat calendars Replace “Day 3 follow-up” with “Signal happened → Play A.” Rule #6: AI assists, humans approve Let AI do the research and draft the openers (1–2 factual lines). You tighten the hypothesis and CTA. Personalization/relevance is proof, not poetry - don't spend hours crafting it. Rule #7: Deliverability and automation Great copy is useless if it never lands. Use Smartlead or Instantly.ai for emails, and take deliverability seriously. Automate the followups on all channels - no, you will not remember to followup with that LinkedIn convo. Don't lie to yourself. (this should also be a rule) Use HeyReach.io to get that LinkedIn game under control and track it. Rule #8: Experiment like an adult One variable at a time. Weekly loops. Measure reply% → positive% → meetings → cost/meeting by segment/source. Scale winners, retire losers fast. Rule #9: Ops-first instrumentation Every LI + email touch writes to CRM. Dashboards by segment × source × channel × ACV. Keep tools in sync (no stale lists). Rule #10: Document EVERYTHING. And that’s a wrap. Next post: our first low-hanging fruit campaign setup. I’ll share the build in public, follow along. Now, execution mode. 😵💫

The hidden reason 90% of outbound campaigns die after 30 days (and it's not what you think). It's not deliverability issues. It's not terrible offers. It's not bad copy. It's that most teams never build feedback loops. They launch a campaign, send it for a month, and when results plateau, they blame the list. Then they start over with new: Copy. Targeting. And sequences. And the cycle repeats itself. Here's what we learned after running outbound for 120+ companies: Your best-performing campaigns are hiding in your current data. You're just not listening to it. At ColdIQ, we treat every reply as intelligence. Prospects' feedback should be leveraged into better campaigns: 1. Tag Every Single Reply We use three categories in Instantly.ai: → Positive (interested, asking questions, booking calls) → Negative (unsubscribes, "not interested," objections) → Neutral (out of office, wrong person, timing issues) But we go deeper. For positive replies, we track: → Which email in the sequence hooked them → Which subject line did they respond to → Which value proposition resonated → Which persona/role they hold For negative replies, we track: → Budget concerns by role → Common objections by industry → And timing pushbacks by company size 2. Analyze Patterns Weekly Every Friday, we pull campaign data from Instantly and Clay. We look for: → Which industries respond best to specific messaging → Which angles get the most positive replies → Which CTAs drive the most meetings Example from last month: CTOs at Series A companies responded 40% better to efficiency messaging than to ROI messaging. So, we built a separate sequence just for that segment. 3. Build Iteration Workflows Based on weekly data, we create new email variations using Claude. But we don't rewrite entire campaigns. We test micro-improvements: → New subject lines for low open rates → Different pain points for cold segments → Alternative CTAs for warm prospects We use Instantly's A/B testing to run these variations against control groups. 4. Create Campaign Evolution Rules When a campaign hits certain thresholds, we automatically evolve it: → If positive reply rate drops below 2% after 500 sends, we test new angles → If objections cluster around budget, we add ROI-focused follow-ups → If timing pushbacks exceed 30%, we build nurture sequences 5. Feed Insights Back Into New Campaigns Every insight gets documented in our Clay database. When we build campaigns for new clients, we start with proven patterns: → Subject lines that work by industry → Pain points that resonate by role → CTAs that convert by company size We're not starting from scratch each time, but building on what already works. The result? Average positive reply rates improve 30-40% between month 1 and month 3. Feedback should guide your strategy. Treat outbound like a conversation where you actually listen and optimize accordingly. Questions? 👇

If 2024 taught us anything about Cold Email, it’s this: 👇 General ICP Outreach isn’t enough to drive results anymore. With deliverability getting tougher every day, there’s only one way to make outbound work: → Intent-Based Targeting Here’s how we do it at SalesCaptain to book 3x more demos ⬇️ Step 1️⃣ Identify High-Intent Triggers The goal? Find prospects showing buying signals. ✅ Website visits – Someone browsing pricing or case studies? (We use tools like RB2B, Leadfeeder, and Maximise.ai). ✅ Competitor research – Tools like Trigify.io reveal when prospects engage with competitor content. ✅ Event attendance – Webinar attendees or industry event participants often explore new solutions. (DM me for a Clay template on this) ✅ Job changes – Platforms like UserGems 💎 notify us when decision-makers start new roles (a prime buying window). ⚡️ Pro Tip: Categorize triggers: → High intent: Pricing page visits → Medium intent: Engaging with case studies This helps prioritize outreach for faster conversions. Step 2️⃣ Layer Intent Data with an ICP Filter Intent data alone isn't enough, you need to ensure the right audience fit. Tools like Clay and Clearbit help us: ✅ Confirm ICP fit using firmographics ✅ Identify the right decision-makers ✅ Validate work emails ✅ Enrich data for personalized messaging ⚡️ Key Insight: Not everyone showing intent fits your ICP. Filter carefully to avoid wasted resources. Step 3️⃣ Hyper-Personalized Outreach Golden Rule: Intent without context is meaningless. Here’s our outreach formula: 👀 Observation: Reference the trigger (e.g., webinar attended, pricing page visit) 📈 Insight: Address a potential pain point tied to that trigger 💡 Solution: Share how you’ve helped similar companies solve this pain 📞 CTA: Suggest an exploratory call or share a free resource ⚡️ Pro Tip: Use tools like Twain to personalize at scale without landing in spam folders. 📊 The Results? Since focusing on intent-based outreach, we’ve seen: ✅ 3x Higher Demo Booking Rates 📈 ✅ 40% Reduction in CPL (focusing on quality over quantity) ✅ Larger Deals in the Pipeline with higher-quality prospects It’s 2025. Let’s build smarter, more profitable campaigns. 💡 Do you use intent signals in your outreach? Drop me a comment below! 👇

I built a workflow that tracks companies engaging with our ads and automatically sends them an email. I use LinkedIn Ads as an intent signal, alongside other inbound-led outbound workflows: → Website visitors (RB2B) → LinkedIn post engagement (Trigify, Teamfluence) Here’s how it works: 1️⃣ Create targeted campaigns ↳ Seniority + Job Function (e.g., Directors+ in Sales) ↳ Job Titles (e.g., Head of Sales, CRO, VP of Sales) 2️⃣ Track engaged companies (using LinkedIn Ads data) ↳ If a company engages with my Director+ Sales campaign, I know it was senior sales leaders interacting. ↳ If a company engages with my Marketing campaign, I know it was Marketing leaders engaging. 3️⃣ Send the company list to Clay for qualification: ↳ Account scoring ↳ CRM lookup 4️⃣ Find the right people in those companies based on the campaign they engaged with: ↳ Identify decision-makers in Sales, Marketing, or other relevant departments. ↳ Cross-check their roles to match the campaign they engaged with. 5️⃣ Verify their emails & contact details 6️⃣ Create AI personalizations 7️⃣ Scale this across multiple campaigns ↳ Replicate the process for different departments. ↳ Each time, knowing exactly who likely engaged based on my targeting. Why this works: → Warm up accounts with ads before reaching out. → Gather real engagement data instead of relying solely on third-party intent signals. → Proactively build outbound campaigns with real-time insights. By adding LinkedIn Ads as an intent signal source, I don’t just wait for interest, I act on it. Comment if you'd like the full-res graphic 👇

I've watched 100+ outbound campaigns FAIL at ColdIQ. Most of the time, it wasn't the copy, timing, or offer. It was THIS... They were aimed at people who were never going to buy anyway. Here's what I mean: Too many companies still run outbound like this: → Pull a lead list from their CRM → Hope and pray something sticks → Fire off 100 cold emails/week to "hit quota" They have no idea why prospects are on their list in the first place. If you're not starting with the right inputs, it doesn't matter how good your cold email is. It's still a shot in the dark. One way to fix this is through intent data: Here are some signal plays we run for ColdIQ and our clients: 1️⃣ First-party intent: Who's visiting your website Not everyone fills out a form, but that doesn't mean they're not interested. We use tools like Instantly.ai and Vector 👻. They track anonymous visitors and identify who's checking out our content, landing pages, or product pages. This gives us a warm list of people who are already aware of us. Even if they haven't raised their hand yet. First-party intent can also come from: → Product usage (Common Room, Pocus) → Social engagement (Teamfluence™, Trigify.io) 2️⃣ Second-party intent: Champion job changes Let's say someone loved your product at their old company. They just switched jobs. Now they're in a new buying position, possibly with budget and urgency. Tools like Common Room and Unify help us track job changes across our network and historical CRM contacts. We can re-engage with a hyper-relevant message, right when they're getting settled in. Second-party intent can also come from: → Review sites (G2, Capterra) → Affinity signals (Crossbeam, WorkSpan) 3️⃣ Third-party intent: Research at scale Most often, you need to go outbound into entirely new territory. That's where third-party data comes in. Pulling insights from: → Hiring trends (LoneScale, Mantiks, PredictLeads) → Tech stack changes (BuiltWith, Similarweb) → Funding rounds (PitchBook, Crunchbase) OR from custom AI agents (Relevance AI, Claygent) We use Clay to build many of these workflows: → Filter for buying signals → Enrich contacts in real-time → Combine multiple data sources → Score and segment dynamically The result? You're increasing your odds of reaching out to the right person, with the right message, at the right time. Better targeting = better reply rates = better pipeline. Whenever your outbound is underperforming, start by reviewing your data strategy. What intent signals are you tracking in your GTM motion right now? 👇

"Someone at Acme Corp showed intent." Cool. Who? "The account is surging." Great. Which human? "They're in-market." Fantastic. What's their name? This is why account-based intent is broken. It's too fluffy. You're calling blind. Emailing ghosts. Chasing shadows. While your actual buyer already moved on to your competitor. Here's the reality: ✗ Businesses don't buy. People do. ✗ Accounts don't have pain. Humans do. ✗ Companies don't raise hands. Individuals do. Yet we're still playing guessing games with "account-level signals." On October 1st, I'm joining Influ2 and my good friend Katie Penner to blow this wide open. We're talking CONTACT-LEVEL intent. - See exactly WHO visited your site - Know WHAT they looked at - Understand WHEN they're actually ready - Plus (and this blew my mind) AI Search intent as well!? What?! But here's the kicker: The signal means nothing if you blow the outreach. "Hey, I saw you were on our website" = Delete "I noticed your company showing intent" = Trash "Looks like you're researching solutions" = Ghost That's why I'm sharing my KPICC framework for email/LI/Video/Voicemail: - **K**now - here's what I know about you/company/persona (not just the trigger) - **P**roblem their persona/company may be dealing with - **I**mpact of that problem, what it actually causes.   - **C**TD - Connect The Dots why you're reaching out. - **C**TA - Call to Agreement (not always action) Real playbooks. Real examples. Real results. Teams using contact-level signals see 3-4x higher conversion. Not because the data is better (though it is). Because they finally know WHO to call and WHAT to say. No theory. No fluff. Just tactical plays you can steal. Stop guessing. Start knowing. See ya'll there... https://lnkd.in/eCVNc6sE