Interview Them: the full GTM Ventures system (audiobook) | Charles Cormier

Interview Them by Charles Cormier. This is Charles, the AI version of me. This is the full GTM Ventures system, told as a story, from the first invite to the client who sends you the next one. Let's go! Read this first. I've been sending cold emails for a living since 2020. I've interviewed more than 3,000 founders and CEOs on a podcast at 1.15 a day. About half of my customers came from those conversations. This is the book about how that works. All of it, from the first list to the client who sends you the next one. It's also a story. The system in this book wasn't designed. It was assembled out of things that broke: a business that collapsed in a pandemic, an agency of hundreds of people that never scaled, a cold email channel that stopped working, and a week in September of 2026 when my replies fell off a cliff. Every rule in here exists because I paid for it. So the book has 2 parts that run side by side. The story is what happened to me, including the dumb parts. The play is what to do about it, step by step. The first part tells you where the idea came from. The second part is the system in 9 steps. The third part is the machine of agents that runs it while I'm asleep, what broke, and how to run it yourself or have us run it. If you're thinking about working with us at GTM Ventures, this is exactly what we build and run for you. If you already work with us, this is the playbook behind your campaigns, so you know what each piece is for. Two promises: every number in this book comes from my own accounts and my own writing, and where a number is an estimate, I say so. And every client who isn't already public on our website is anonymous. This is a living book. Every time I learn something new about the system, it goes in here. The version you're reading is the latest. Charles. Stop booking impressive guests. Start booking the 20 people you actually want as clients. Part 1: The Story. Chapter 1: The Kitchen Table. I started my business activities in 2016, with dropshipping. The agency work came around 2017 and 2018. The first thing that really worked was a list, and I'll start there. I was young, back in my parents' house, and I had two choices: sell something or keep being miserable. I'd left a marketing job. I'd quit an agency I started with a partner. I had a laptop, a phone, and the kitchen table in Boucherville. So I did the only thing I knew how to do. I pulled a list of every real estate agent in Quebec, dropped it into a Google Sheet, and made a color code. Red meant no, yellow was voicemail, green was send me the info, purple was closed. I called them, one after the other. I'd started my career doing 200 cold calls a day for months, which teaches you what a no is worth. It's worth almost nothing. I was already past 1,000 names on the A's. By the time I got to the B's I called a man with a 418 number and he said yes. A $4,000 website. Then search ads, then more. He turned into one of the best clients I ever had. That's where I learned the first thing I still believe: a list is worth a million. Not the data, the list. The right names, in a sheet, with someone willing to call every one of them. The Play: The Lesson of the First List. Most of what I do today is that Google Sheet, with better tools. Pick the group of people who need what you sell. Put every name in one place, contact them in order, track what happens, and never stop until you finish the list. I got my first big deal that way. Years later, a list built the same way got me Jordan Harbinger, John Lee Dumas, and Andrew Warner on my podcast. Then I made the mistake that taught me the second thing. For 2 years, I served real estate agents. They were easy to reach and they paid, and I got to $30,000 a month in recurring recurring revenue while I was still very young, which wasn't bad. I wasn't inspired by the profession. I'd seen money in a niche and gone in. It scraped me from the inside. Then in 2018, those same clients turned on me with chargebacks, and it burned the business. In 2019, the government shut down my call center. In 2020, COVID brought everything to a halt, and I lost most of my clients, because real estate agents couldn't physically visit houses. 3 years, 3 disasters. I don't tell you this for sympathy. I tell you because it's the real reason this book is built the way it is. The lesson I took away: have multiple businesses and a plan B, C, D, and E set up because something always hits the fan. And I learned something about clients that I use every week. Clients who complain about a $200 product will charge it back and almost kill your business. I know, because mine did. Pick clients who are as good as the best version of yourself. Cash was the other thing those years taught me. When money was tight and I was traveling with a team, I blocked whole days in my calendar and called them money days. On a money day, I did one thing: sell. No meetings about meetings, no planning. I cold-called or I emailed until cash came in. It always did. It still works. One December, after a few slow weeks with no closes, I got annoyed, ran a holiday promo at roughly half price, added more cold email, and closed $12,000 in a week. I've noticed that there has never been a December 20th to 30th in which I didn't generate $20,000. The play. The 2 rules I took out of those years: choose your client before you choose your channel. If your client is wrong, nothing you do will work. Never depend on one thing, one niche, one platform, one inbox, one customer. Each of them will let you down once. I'll show you in later chapters how both rules turn into specific settings, how I pick guests like I pick clients, and why my system never sends from one domain. Chapter 2, The Year Cold Email Broke. When COVID hit, I was doing lists and prospecting for real estate agents. Agents couldn't physically visit houses. My revenue stopped for weeks, and I knew I had to pivot hard and fast. LinkedIn outreach was the obvious move, and it was already broken. LinkedIn was jailing accounts that connected with more than 200 people a day. The channel I'd built a business on was closing. Then I found Apollo. It cost about $50 a month for 10,000 contacts, and it connected to an email sender—contacts and sending under one roof. I was hooked almost immediately. I got 5 meetings a week out of it, which was impressive for the time, and then more. Within a few years, I was doing 50 and more. I'd spent a long time afraid of selling to English Canada and the US because of my French Quebec accent. At best, my accent is cute. In early 2020, I expanded to the rest of Canada, and in 2021 to the US. In 2021, I started podcasting. The first plan was a SaaS sales podcast, somewhere between Saastra and My First Million. I sent my first 100 invites and wrote that I'd probably open my mailbox like a candy store. So by 2022, I had a cold email agency doing well, a podcast, a team, and an idea that cold email was the closest thing to a modern goldmine. Then it started to decline. The play, what I saw in 2022. Open rates dropped, spam filters got tougher, only compelling offers made it to the inbox. Prospects were getting 20 emails a day from people selling the same thing, but the podcast invites kept getting through. Sales emails were getting ignored. Podcast invites kept getting through, and they still do. For the sales skeptical CEOs, the invite is almost the only outreach they'll say yes to. So I merged them. I used cold email as the delivery system and the podcast as the offer. The first month, it was magic. I could get 10 CEO meetings a day and close about one in 10 into a client. For about four months in 2023, I ran it flat out. Fifteen CEOs a day. I wrote at the time that I got three to five refusals daily and twelve to fifteen yeses, and that I didn't care about the refusals because each guest also promotes the episode to their own audience. I stopped reading books. It was like getting thousands of dollars a day of free consulting because I could ask fifteen CEOs a day exactly the questions I needed answered for my own business. I recorded with the people I'd always wanted to meet. A list I built of other podcasts' guests got me Jordan Harbinger, John Lee Dumas, and Andrew Warner. Andrew Warner grilled me for 10 pieces of feedback on my Zoom background and my 65 LinkedIn experiences, called me a jack of all trades, and a month later came back for round 2. It also nearly broke me. There was no day without 10 meetings. Deep work became a theoretical concept. After 4 months, I cut the number by half because batteries need to be Recharged. And I learned something I didn't want to learn. I talked to perhaps 1% of the people I interviewed and learned something new. Years later, after more than 3,000 founders, I still believe it. About 10% of founders are really successful, and about half can't explain their own playbook. Maybe 1% can say clearly what they did. The conversations were worth it for a different reason. The most successful people were the hardest to get, and they were slow to say yes to a podcast. So I changed the order for them: sales call first, podcast invite after. I still use that order for some of them. During those same years, I made a different bet, and it failed. I believed that if I hired enough people, I could scale. I hired and hired. By the time I counted, it was more than a thousand people through job boards, freelance platforms, and agencies. In 2026, I put the number at over 2,000, and zero of them are still with me. About 90% were negative ROI. One year, I looked at my team and wrote a post about feeling pride. I gave everyone a raise. A few months later, I'd fired about a fifth of the team for performance and culture. I wrote that I'm not the white-gloved, soft kind of leader, and that some of my people were scared to talk to me. We lost an intern, and people cried. I outsourced way too much to people who were never going to care as much as I did. I estimate it cost me somewhere between one and ten million dollars a year in missed revenue over seven years. Then, in September 2023, I let go of my very last employee. I called it my full AI metamorphosis. The feeling was relief. The numbers were harsh. I went from thirty thousand dollars a month in recurring revenue to zero, and back to twenty thousand dollars in three months, rebuilding with me and AI. I paid a chief of staff to write procedures, then realized on my motorcycle that I could record my calls and feed them to ChatGPT. It took me 6 years and OpenAI to understand that employees wouldn't scale. That is why the system in this book is built around agents. I didn't choose it because it was fashionable. I chose it because I had failed at the alternative many times. What'd I tell you about the people problem? Don't hire to fix a system that doesn't work yet. More people multiply a broken process. Build the funnel first and let it produce the team. Today I hire from past clients and from sales call prospects who can't buy— don't buy, join. Pay for outcomes you can measure. Full commission reps don't stay unless the results are consistent. Salespeople I recruited as resellers delivered nothing. From Agency to Partner. For years, I sold lead generation as a service. It worked. It still does. But there came a point where I looked around and saw 50,000 competitors. My ideal customers were getting 20 emails a day from people selling the same thing, and I was one of them. I also noticed that I was selling time. A client paid me a retainer, I generated leads, and whether the client closed them or not, I got paid the same. I sent tens of thousands of leads that founders never closed. Two things changed my mind. The first was equity. On my 30th birthday, I wrote that I had started taking shares in clients' companies—one to five percent vested. Then I made it a system—a commission and equity structure for every client—as a way of filtering for the people serious enough to share the upside. With my favorite clients, I work full equity after three months. The second was Sunny. Sunny Ray was first a client. I generated 300+ leads for 2 of his companies, one in AI and one in Bitcoin. I do something with my clients that I call reversing the filter. I check whether the client's product is good enough that the system should work, because the machine only works when there's product-market fit. Sunny's product passed. We tried a bunch of things, and it worked. Then he started referring me clients, a lot of them. One day I said, in effect, I like you, you like me, why not team up? Come in as a co-founder on this new business I'm launching, Razrocket. He accepted. I once posted a joke that explains it. A technical person says, that guy can't sell for shit. A non-technical person says, that guy can't build for shit. Complete, don't compete. Sunny and I share about 95% of our views, and we debate the rest. Razrocket does one thing: we run the same cold email and podcast machine, but for founders who are raising money, and the audience is investors. Some of what it produced is public on the site: $16 million in term sheets in 28 days for a robotics company, 300+ investor leads for another, The play. The rebrand that wasn't cosmetic. In December 2025, my agency became GTM Ventures. I wrote the same day that it wasn't cosmetic. It was a hard pivot into partnering with the top 0.1% of builders on performance and rev share. No more agency model, no more selling time. The pitch to a client became equity, rev share, and long-term alignment, not retainers. The line I wrote was that we're structuring ourselves to compound alongside elite founders instead of acting as an external vendor. In early 2026, with Sunny in, we relaunched something I'd tried years earlier and failed— closing as a service. The problem back then was logistics, not the market. This time, I handle lead generation and system design, and Sunny handles closing and deal orchestration. I'm elite at lead gen. Which is my craft. At closing, I'm about a 7 out of 10. The line we use now, we're not here to help you sell, we're here to own the outcome. The play, what a partner is. Partners are the people who are in the machine with you, not the people who buy from it. Clients become partners. A client who gets results and who I like gets an offer to become a partner. Partners bring clients, they get a share of what they close. Everyone's incentives point at the same number. Revenue that lands. I'm not against retainers. A lot of GTM Ventures clients still pay a monthly fee. But the fastest way I know to find out whether someone will do the work is to share the upside. Around the same time, I started making things that weren't lead gen. A live pitch show called Spark Tank. 10 startups, 5 investors, 90 minutes. Each founder gets 3 minutes and each investor says in or out in 1 minute. With a reason. It went live in February 2026. I didn't think of it as another business. I'd said it plainly: I don't start things, I align interests. Investors want deal flow, founders want capital. Almost nobody is lined up. A live show is a room where they are. Then, a week of masterminds in Oaxaca. 10 founders, no sponsors and no stage, with the partners: Sunny, Amir, Emiliano, and others. Fire pits, poker, cigars, founder movies, and mezcal. Everything I run today comes from the same 4 ideas. Invite, don't pitch. Interview them. The conversation is the asset. Give first. The package, the introductions. Share the upside. Partners, not vendors. The rest of this book is how to build the machine that runs all 4. Chapter 4, Why an Invite Works. If you cold DM someone, their guard is up. If you invite them onto a podcast, their guard drops immediately. You're no longer selling, you're giving them a platform. That one difference changes everything. Most cold outreach is transactional. Buy my thing, hop on a call, can I have 15 minutes? That's why it gets ignored. The biggest bottleneck in B2B outbound isn't sending, it's what you're asking for. A podcast invite is the best offer I know in B2B. It feels collaborative, high status, value first, socially safe, and ego rewarding. Instead of give me, it's let me showcase you. Stop thinking you're the next Joe Rogan and that you'll make a fortune from podcast ads. Put lead gen in the podcast equation. You don't need an audience. The power is in the conversation, not the clout. The play. What changes when the ask is, come on my show? Volume. You get a pipeline that feels almost limitless, even in a crowded niche, even when your offer isn't perfect yet. Deliverability. You're not pitching, you're inviting. Invites don't trip spam alarms the way sales emails do. Signal. Every interview is a discovery call in disguise. You hear their pain, their language, their objections, and how they decide While they're in a good mood. Trust. If you want someone to trust you, spend time with them. 45 minutes on a show beats 10 cold emails. Content. Every conversation is also an episode, clips, and posts. One system, multiple outputs. That's leverage. The play. The intelligence advantage. The hidden advantage isn't even the audience. It's the intelligence. Pick a very specific customer, say fintech founders, and interview them relentlessly. Use the conversations to understand their bottlenecks, fears, hiring problems, go-to-market issues, revenue leaks, and pressure. Patterns show up fast. The same frustrations, the same objections, the same hidden pain. Now you're not guessing what to sell. The market is telling you. Most founders stop talking to customers once they think they found product-market fit. That's exactly where companies start slowly dying, because markets and pain keep moving. My competitors talked to their customers 3 years ago. I talked to mine this morning. That gap is the whole game. The play. A podcast is a discovery call in disguise. A great sales discovery call is all about them. You ask great questions and help them think through their problem. You act like a consultant. A great podcast interview is exactly the same. You ask about their goals, how they get customers, what's not working. You add value and you show how you think. I like to picture it as a glass of water. Every conversation, every follow-up, every useful post pours a bit more into their glass. When it overflows, the overflow is what they pay you. The podcast is the first pour. It's the door opener, the zero to one. The play. What it isn't. The— it isn't a media company. Don't charge guests to appear. It caps you at a small business and repels the people you want. Use the podcast as the first step to your paid offer, never as the product. My sequence: podcast, then community or mastermind, then paid offer, then the client becomes an affiliate. None of it works once. It compounds over years. That's why it looks unfair from the outside. Part 2: The System in 9 Steps. Chapter 5: The Map, 9 Steps. Here is the whole thing on one page before I walk you through it. Everything I run at GTM Ventures, for me and for clients, is 9 steps in a row. Each step has a job, an owner, a number I watch, and a way it breaks. If you remember nothing else from this book, Remember the order. List. Decide who you want as clients. Build the list of exactly those people, verified. Infrastructure. Separate domains and mailboxes, warmed, monitored every day. Campaigns. Invite them onto your show. Short, plain, no links, no pitch. Replies and booking. Answer fast, in your voice. A short form, a score, a 15-minute pre-interview. The interview. 30 minutes. They talk, you learn. A discovery call in disguise. The package. Give them the episode, the clips, the post, the intro. Free. Before you ask for anything. The close. Ask. Send the proposal inside 24 hours. Put a date on it. The loop. Follow up for a year. Warm people, watched for signals, never dropped. The compound. Guests bring guests, clients become partners. The next list gets better than the last. That's the front door. A stranger gets an invite, becomes a guest, becomes a friend, becomes a client, and sends you the next stranger. Here are the numbers that sit behind it, all from my own accounts, so you know what scale I'm talking about. More than 3,000 founders and CEOs interviewed at 1.15 a day, about 100 podcast meetings a week at the peak. One of my invite campaigns, 344 replies from 61,000 sends. My warm re-engagement campaigns, 3% to 7% reply. My cold ones on a bad day, 0.1%. About 3,800 inboxes across 83 domains at the biggest point, running in 4 Instantly workspaces, mine and my partners'. More than 60 scheduled agents running pieces of this while I sleep. I'll tell you what each of those means, and what they cost me to find out. The Play. How to read this book. Each chapter has a story and a play. The story is what happened to me, including the dumb parts. The play is what to do about it. If you're in a hurry, read the plays. If you want to understand why the plays look the way they do, read the stories. The stories are where the system actually comes from. Chapter six, step one, the list. Everything starts with a name on a spreadsheet, and most people get this step wrong in the first five minutes. They open a database and type in the job titles they like. They get forty thousand rows. They feel rich. Then they send, and nothing happens, and they blame the copy. I've built a lot of lists: founders hiring SDRs, franchise sales reps. Ivy League alumni who run companies, PR agency owners, M&A brokers in Texas, investors who write checks into Bitcoin. Every one of them taught me the same 2 things. The first thing: the list is the offer. If the guest profile is wrong, no copy on earth saves it. The second: the list is an asset, not a purchase. You pay for a contact once; you shouldn't pay for it again. The play, pick guests like you pick clients. Write the profile of your best client, not your average one, the best one, the one who pays well, stays, refers, and is a joy on a call. Then write the profile of the person at that company who says yes to the podcast and 6 months later signs the contract. That's your guest. One rule I learned the hard way, count the buyers, don't guess the market. When I score a lead, the size of their market is a headcount of reachable buyers, not a dollar figure. If someone sells to 300 possible customers in the whole country, I can't help them, however nice they are, because I can't fill a campaign from 300 people. Under about 1,000 reachable buyers, I send them somewhere else. I'd rather tell someone the truth on the first call than take their money and burn their list. The same count works on your side. A client list should be thousands of people, not a few hundred. 245 contacts is not a campaign. It's a hobby. The play. Where the names come from. I use everything. And I use it in a strict order: cheapest first. What I already own. Every export, every past campaign, every guest, every person who ever replied. I keep all of it in one local database. At last count, it held about 4.3 million rows from 1,683 files, with 1.39 million email addresses, plus every verification result I've ever paid for. Free search. Some data tools let you search for free and only charge when you unlock the email. Searching by school, title, and country costs nothing. I once pulled 37,000 Harvard-educated founders and CEOs this way before spending a cent. Match, don't buy. For most of those 37,000, I already had an email or could build one from the company's pattern. Roughly 12,000 matched for free. Unlocking a missing one costs 2 cents. Verify only what's old. This is the part people skip, and it's where most of the money leaks. The play: freshness lights. Every email address in my database carries a light. Green: verified in the last 60 days. Send. Yellow, 61 to 180 days. Fine for warm lists. Recheck before a big cold send. Red, older than 180 days or never verified. Verify first. Dead. It bounced or came back invalid after the last time it was good. Never send ever. When I counted, I had about 89,000 green, 85,000 yellow, 958,000 red, and 17,400 dead. Nearly a million emails that looked usable and weren't safe to send on. That's the difference between a list and a database. The check pays for itself. Across 171,896 contacts emailed from my main workspace, 257 bounced. That's 0.16%. Of those, 177 had never been verified at all. When you verify, Bounces almost disappear. When you don't, they're the main reason your domains die. The play: protect the people on the list. Keep a do not contact file and check it before every import. When someone asks to be removed, they go in the file, and the file goes through every campaign, every tool, and every sheet. One founder emailed asking to be erased. He went on the list the same day. It costs you nothing. It's the law in a lot of places, and the people who ask are exactly the people who will remember how you handled it. The play, build it in layers. You don't need the whole list on day 1. You need 3 layers. Layer 1, warm. People who already know you, past guests, people who replied before, people who said yes and never booked. This is your first campaign, and it's the one that works best. Warm lists reply at 3% to 7%. Layer 2, adjacent. People one step from your network. The same schools, the same investors, the same communities. Layer 3, cold. Everyone else who fits the profile, verified in thousands. Launch Layer 1 before Layer 3 exists. You'll learn more from 300 warm replies than from 30,000 cold sends. Chapter 7, Step 2, Infrastructure. On September 23rd, 2026, my replies fell off a cliff. I had four instantly workspaces running, mine and three partners, around 3,800 inboxes on 83 domains. The week before, one of the workspaces had tripled its daily volume from 1,900 emails a day to 5,600. By September 25th, replies in all four workspaces had collapsed. In my own, replies had been running 40 to 76 a day the week before. A couple of weeks later, they were somewhere between zero and ten. I did what everybody does. I looked at the warm-up score. It read 100% on every single inbox. That number is useless. I found that out the expensive way. Warm-up scores mostly measure warm-up traffic, mail sent between boxes that are all trying to look friendly. They say nothing about whether a stranger at a bank will ever see your email. So I audited the real things. Here's what I found. 2 of our domains were on the Spamhaus domain blocklist. 10 more were flagged by another list. Every inbox on them was now sending mail that got filtered before a human ever saw it. 6 domains had no SPF and no MX records at all. They'd been sending for weeks with the doors unlocked. 35 domains were shared across 2 to 4 workspaces, so one burned domain hurt several accounts at once. Almost all of the bounces in one workspace, 782 of 812, came from 7 domains that lived on one self-hosted mail server with a broken reverse DNS record. The leads were fine. The server was the problem. The same 25 client inboxes were connected into all 4 workspaces at once, each one sending for 4 different accounts. On October 3rd, I paused 477 inboxes, every one on a blocklisted or broken domain. My daily sends dropped from about 50 5,100 to about 900. It felt like cutting off my own arm. Then, on October 11th, it got worse. The hosting company that ran about 120 of my domains suspended the account. Hundreds of mailboxes on one shared server were all being polled by the sending tool at the same moment. That tripped their limits, and they reset every password and labeled the lot cold email outreach. About 850 inboxes went dark overnight. Their message was clear: a shared hosting plan was never built to carry that many cold email inboxes. They lifted the suspension the next day. I'm putting inboxes back at 100 a day, at low limits, and I'm not rebuilding on a shared plan again. The Play: What the Infrastructure Actually Is. Think of it as a small factory with 4 parts. Domains: Extra domains that look like yours but aren't yours. Never send cold email from your main domain. If a sending domain burns, you lose a spare, not your company. Mailboxes. Several per domain, each with a real person's name and a real photo. Start each one at about 10 emails a day and add a little every day. The sending tool. I use Instantly. It rotates inboxes, spaces the sends, tracks replies, and keeps the lead lists. Monitoring. This is the one everybody skips, and it's the one that saves you. 5 rules came out of the September 23rd week. Never trust warm-up scores. Run a seed test instead. Send from each domain to real Gmail and Outlook accounts you own and see which folder it lands in. Check block lists every day. Not weekly. Daily. One domain, one workspace. Never connect the same inbox into several accounts. Plain text, no links. More on that in step 3. Don't put hundreds of cold email inboxes on one shared hosting plan. Use a provider built for it, and keep the domains at the registrar. The play. The thing that changed my reply rate most? Different inboxes receive your mail differently. I exported 89,661 leads and split the reply rates by where the recipient's email lives. Google Workspace recipients, 6.15% reply. Microsoft recipients, 1.81%. Everyone else, 1.28%. Same copy, same list, same week. A third of the difference between this campaign works and this campaign is dead was whose inbox it was going to. The fix wasn't clever. I split big lists into lanes. The Google lane gets higher volume. The Microsoft and corporate lane gets low volume and its own testing. My new campaigns are now about 70% Google addresses. Also, learn from the silent failures. When I tested which domain endings Gmail trusts, every .com landed in the primary inbox, and the .xyz domains were silently dropped. I took them out of every new campaign. The play? The guard. I have a scheduled agent that runs every morning at 8. It checks every inbox in every workspace. Bounce rate at 4% or higher? Auto-pause the inbox. Domain appears on a block list? Auto-pause every inbox on it. A list under 0.3% reply after 500 sends? Flag it. A/B test with a clear winner. Tell me which. I built it after the collapse. I wish I'd built it before. The play. The real limit is a number on your plan. One more lesson, because it surprised me. In early October, my main workspace hit its plan limit of 75,000 contacts. I assumed I needed to upgrade. I didn't. When I looked, 57,000 of those 75,000 contacts were sitting in 4 capped cold campaigns replying at 0.1% to 0.25%. They were using up the room while producing almost nothing. The fix was housekeeping. Delete finished contacts from the weak campaigns, protect the campaigns that work, the warm re-engagement ones, and refill automatically with fresh verified leads. Your plan limit is not a growth problem. It's a signal that you're paying rent on dead lists. Chapter 8. Step 3, the campaigns. The best campaign I ever ran had the dumbest name. You've been on my podcast. It went to people who had already been guests. They knew me. I wasn't selling them anything. I was saying hello and asking if they wanted to come back or send someone good. In that campaign, 92% of the people who replied yes ended up on my calendar. Across all my campaigns, the average was 28%. That gap is the whole lesson of this step. When I counted, I had 1,068 positive replies in my main workspace, Only 301 turned into a booking. 756 said yes, and then I never saw them on the calendar. The biggest leak was a direct sales campaign: 485 positive replies, 15% booked. I had been staring at reply rate for 2 years. The number that mattered was reply-to-booking. So when I tell you what I do in campaigns, I'm going to tell you in the order of what leaks the most money. The play. The invite is the product. Cold email breaks when you ask for something. A podcast invite doesn't ask; it gives. Buy my thing is a request. The reader has to decide. I'd like to interview you is a gift. The reader has to feel flattered or at least curious. When I first merged the two, my first campaigns booked about ten CEO meetings a day. The invites aren't tricks. I do interview them. I do publish it. The pitch comes later and only if there's a fit. Out of 61,000 sends in one of my invite campaigns, 344 people replied. That's a small number on paper. But it's 344 people raising a hand to talk, and I didn't have to convince any of them to open the door. The play? The rules that survived. I've broken every one of these at least once. No links. Anywhere. In any step. Not the booking link, not the podcast link, not your website. Links are what spam filters read. I had banned them, and a booking link still ended up in step 2 of a client's campaign. We found it and stripped it. The call to action is a reply. Open to it, or can I send you more info? One plain subject, 2 to 5 words, lowercase, like a text to a friend. No spintax, no variables. No title case, no emojis. May call. Part 2. The 95% nobody monitors. Only 2 variables: first name and company name. If I want to mention a show, a topic, or a hook, I write it directly into the campaign for that list. Custom variables show up as raw text in some tools when a field is empty, and that is how you send hi hook to a CEO. Short. 45 to 80 words for the first email. One idea, one question, plain text. No images, no signature banners. Sign with a name: Charles Cormier. Not a team, not a company. 3 steps, not 7. First email, a nudge 3 days later, a short goodbye 4 days after that, last one from me. If the timing's off, no worries. Test inside one campaign. If you want to compare 2 versions, make them variants A and B of the same step in the same campaign. 2 separate campaigns share nothing, so you can't tell which version won. Make the words match the list. A campaign went out to CEOs of Texas companies with copy that said business owners. I caught it and told my team, catch these things. Before launching anything, read every step against the list. The title, the state, the company size. It costs 2 minutes. The play, what the email looks like. This is the shape. It's an example of the structure, not a template to paste. Yo, Dana, and saw your team just opened a second office. Hardest part of that is usually keeping the first clients quality while you hire fast. And I host Founder Wisdom, 30 minutes with founders on how they get customers. Would love to have you on to talk about how you did it. And can I send you more info? And Charles, that's it. One real detail about them, one sentence on why it's hard so they know you read it, one sentence on the show, one question. If I don't have a real detail about someone, I don't make one up. I write, hey Donna, I host Founder Wisdom, 30 minutes with founders on how they get customers. Would love to have you on. Can I send you more info? It's less impressive, and it works better than a fake compliment. The play: identity beats cleverness. The campaigns that performed best for me were the ones that named where someone came from. You've been at Harvard. You're an ex-Goldman. They hit around 75% opens and 18% replies at scale. Mentioning that you're targeting ex-Amazon people does more work than most of your competition does in a whole sequence. People have a strong tie to where they came from. The same goes for a city in the subject line or a venture fund that backed them. You confirmed that fund invested in them, and now they know you're legit. I'm giving you one honest warning about numbers. The best campaigns I ever ran were small and early. My honest average across everything at scale is 1% to 5% reply and about 7% on a very good day. When you see someone claim 40% reply rates, that's a sales page. The play: invite the people who've been invited before. My favorite list is people who have already been guests on other shows. A founder who has done 20VC or Lenny's or My First Million already says yes to podcasts. They know the format. They don't need convincing that it's worth an hour. I built a campaign per show: Pod, 20VC, Pod, Latent Space. Pod, Invest Like the Best. Same invite, one line changed. The warmest cold list you can build is people who said yes to someone else. The Play: The Personalized Lane. I run a second kind of campaign beside the bulk one. A separate set of inboxes, a separate workspace, and every email is written for one person. A research step reads their company page, their latest round, what they posted, and if I've spoken to them before. My notes. An AI drafts the email from that. A checker validates it with hard rules: no invented numbers, no invented news, no fake compliments, no more than one question. One rule I'm proud of: if the email opens with a news item, that headline must be a real one from the last 10 days, from a named outlet, and the checker verifies that the headline actually exists. Fresh news from their own industry shows you know their world. Invented news shows you don't. Volume here is small, a few dozen a day. It's new enough that I can't give you a reply rate yet, and I'll update this page when I can. The point is not to replace bulk. It's to learn what good looks like so the bulk gets better. The play. For enterprise, use LinkedIn first. Big companies have email filters that treat cold email like a fire. For a list of about 1,000 enterprise accounts, my approach is the opposite. LinkedIn first, email only as a follow-up. The cap is around 20 connection requests a day per account, so I use first-degree direct messages, which are free, and I spread the work across several seats. It's slower. It's the only door that opens. The play: seed every campaign. Before any campaign goes live, I add a few friendly addresses— myself, my partners, and a teammate As leads in it. That way, the people running the system see exactly what a prospect sees. You'd be amazed what you catch. A broken signature, a wrong name, a line that sounds like a robot. The play. Launching is a human decision. My agents build campaigns, load leads, run the checks, and leave everything paused. A person presses go. This is my rule for my own campaigns and an absolute rule for clients. It has caught the wrong copy before it reached the right people. Chapter 9: Step 4: Replies and Booking. One of the costliest mistakes in my whole system was a robot being nice to the wrong person. I had an AI agent answering replies. It was fast, it was polite, and it never slept. A founder replied to a podcast invite. The agent answered with a calendar link. The wrong calendar link. It was the link for a sales call with me, not a podcast pre-interview. The founder booked it, thinking he was booking the podcast. I found a second one like it soon after. Another lead told me my agent was stuck in a loop, so I paused the fully automatic agent. Replies are now drafted in my voice from the person's own words, and anything that matters waits for me. It's slower. It's also the reason the people who book are the people who wanted to. AI can reply. But a reply is where the relationship starts, and a wrong first answer is hard to undo. The bot wasn't always the villain. In November 2025, I took a last-minute 5-day trip with no laptop. When I came back, my AI replier had booked 20 meetings from campaigns I'd launched 2 or 3 weeks earlier. I was watching The Lion King in Mexico City at the time. That's the seduction. It works so well when it works that you stop looking. The play. Answer inside hours. A yes is perishable. Answer the same day, ideally inside 2 hours. The reply has 3 parts: a thank you in one line, what the show is, how long it takes, and who's been on. A single next step. Ask the question first: Wanna pick a time? Send the link only after they say yes. Never send the booking link in the first email. Never send 2 links. The play: know what people say and answer the real objection. I read 15,180 replies from my own campaigns and sorted the objections. Here's what shows up and what I say. Is this pay-to-play? The most common reply to a podcast invite. No, no cost, no fees, ever. Say it in the first line. Send me examples. Who watches this? Send the channel and 2 episodes in their niche. Nothing else. Not the right person. Thank them and ask who is. Is this AI? 119 replies asked, in one way or another. A real person reads every reply. I host the show myself. Point to episodes. Too expensive. 471 replies. Spam. 384. Not now. 261. These are the sales objections, and they show up when the email is a pitch. A podcast invite gets a shorter, friendlier set. I keep this list in one document, and every card on my calendar pulls the top objections from it. The play. The booking form. The booking page asks a few questions. They're the cheapest sales qualification I've ever found, because the guest answers them while they're excited. Mine ask about their company, their revenue band, whether they have a sales team, whether they're hiring, whether they're raising money, And where their leads come from today. Most scheduling tools cap you around 10 questions, and I use them all. Ask the questions that matter to the interview and the follow-up, not the ones you wish you knew. Every extra field loses a few bookings. The play: a score on every call. For every booking, an agent reads the form and scores it out of 100. It adds a card to my calendar next to the call. Only I can see it. The card has traffic lights for the things that decide whether I can help: how many buyers they can reach, counted as a number of people; the size of a typical sale; whether their market can be listed; whether they have budget; whether they have a sales team to take the leads; how urgent it is. It also has my estimate of their yearly revenue, the package that would fit them, the 5 objections they're most likely to raise, And a case study from a client who looks like them. 2 rules I added after a few weeks: the card never cancels anyone. If a lead scores low, the card tells me, and I decide. A system that rejects people automatically will, sooner or later, reject someone it shouldn't. Every low score comes with a way out. If I can't help someone, I point them to someone who can: a partner's show, a different kind of program, A person I know. A no that comes with an introduction is worth more than a yes that disappoints. Know where the booking came from? For a long time, I could tell you how many calls I had. I couldn't tell you which campaign, which inbox, or which email produced them. Calendly bookings had no source data at all. Now each card includes a source line: the workspace, the campaign, the inbox that sent the first email, The date of the first touch, the reply, and who sent the booking link. Source is super important. It's the only way to know what to do more of. The play: close the leak between yes and booked. This is where the money was hiding. I matched every positive reply in my main workspace against my actual calendar. Of 1,068 people who said yes, 301 booked. The other 756 vanished. Not all 756 are gone for good. Some people book later, and some of my partners' campaigns send to their own calendars, which my count can't see. But a lot of them were real. So I built a daily job. It pulls the yeses from Instantly, pulls the bookings from Calendly, matches them, and marks the booked ones. Then it drafts a follow-up for the good fit people who never booked. At the first pass, it found 154 real yeses since July and 85 follow-ups ready to send. The follow-up is short and human. It refers to what they actually said. It doesn't say just following up. The play: no-shows. A podcast funnel lives or dies on show rates. Remind them the day before, an hour before, and a few minutes before. Use email and text. Keep the confirmation message to one purpose. I had to fix a set of pre-call emails that were duplicated and confusing. People weren't sure what they had booked. If someone cancels without a reason, email them. Most are rescheduling, and a few are lost. The few are worth a note. The play. The 15-minute pre-interview. Before the recording, a teammate does a 15-minute video call with the guest. It's friendly. It picks the topic, tests the audio, explains how the episode will be published, and quietly answers the question. Is this person a fit for anything we do? A lot of deals start right here, and nobody pitched anything. Chapter 10, Step 5: The Interview. A podcast interview is the cheapest high-trust meeting I know how to get. At the peak, I recorded up to 15 a day. I'd start a call with a stranger, and 30 minutes later, we were 2 people who knew a lot about each other. They told me how they got their first customer, what keeps them awake, which hire went wrong. Nobody does that on a sales call. On a sales call, people protect themselves. On a show, they perform for an audience, and the audience makes them generous. That's the disguise. You're not tricking anyone. You really are interviewing them, and the episode really will go live. But while you do, you also learn the 3 things a sales call is supposed to learn. What hurts, what they've tried, and how they decide. After enough of these, you stop guessing what the market wants. You talk to the market this morning. Some of those conversations are the reason I keep doing it. Two of my most liked posts ever are about one guest, a community builder named Lloyd Lobo. I sat back for an hour and didn't interrupt him because he had a story, a voice, and something to say. When people ask me why I interview 15 CEOs a day, that episode was my answer. Another time, I formed a bond with a guest in under 30 minutes by agreeing on air to do a 52-hour endurance challenge together. None of those was a pitch; all of them were the start of something. The play before the call. Prepare for 20 minutes; almost nobody does, and that's your edge. Read their last few posts. Read the booking form answers. The guest already told you what hurts. Note something recent about their company. Write one question nobody else would ask them. For every guest, there's also a card on my calendar with a score, a source, their likely objections, and a case study that fits. I read it for 2 minutes. It tells me whether this is a guest, a prospect, or both. The play. The questions that sit in the middle. Before each recording, my scheduling tool emails the guest a short set of questions. One set asks 7 questions about AI. Another asks questions I like asking, about the stupidest decision they ever made, the day they got canceled, what they'd rebuild in 90 days. I sent both at once for a while, and some guests found it confusing. One set at a time is enough. Guests answer in writing, and many of them do it well, because it's easier to write a good answer alone than to improvise one live. 61 guests wrote back to the AI questions; 64 answered the others. I compiled those answers into books, with their names and credits, and left out anyone who wasn't comfortable. The books became a reason to talk to them again. You can do something smaller. Ask 2 questions in writing. Use the answers as the opening of the interview. The Play. Running the interview. Start with their story. People love telling it, and it settles the room. Go to the pain. What's the hardest part of the business right now? What have you tried? What didn't work? Those are the questions a good salesperson asks, asked as a host. Ask one more question. Most conversations end one layer too early. The best insight is usually one question deeper than where you stopped. Steer, don't pitch. If their pain is exactly what you solve, you'll both feel it. You don't need to say it. Let them shine. Give them room for their best stories. That's the clip they'll post. Never sell in the episode. The episode is a gift. The ask comes after, and I'll show you how in step 7. The play. Keep the setup boring. I record 30 minutes live on a browser studio that publishes on its own. A decent mic, a quiet room, one camera. A guest needs to know how to join, so I send a short guide 4 days before, and they arrive ready to be on camera. You don't need a studio. You need the same setup every time, so the call is about the person and not the software. The play— what the interview is for. Be clear about this, or you'll waste it. The interview is the first conversation of a relationship, and the product comes later. You get to be useful, you get to be curious, and the guest gets to feel seen. Everything that follows— the package, the ask, the year of follow-ups— builds on how those 30 minutes felt. If a guest leaves the call thinking, that was a good conversation, you're already ahead of every salesperson who emailed them this month. The play. 3 shows, not one. I don't run only one show. My partners and I run several. And every guest gets invited on all of them. That gives each guest more exposure, and it gives each show more guests. It also means that when I meet someone who's the wrong fit for my offer, I can usually send them to the right show, and the introduction is worth more than the sale I didn't make. Chapter 11. Step 6, the package. The episode isn't the gift, the package is. Most hosts thank the guest, send the link when it goes live, and move on to the next call. The guest gave 30 minutes and got a link. So every guest now gets a package. It's free for everyone, and it costs about $5 of agent time in total. I decided that in October of 2026, free for every guest, extra attention for a few, and the pitch comes after the value. This is the part of the system I'm most pleased with, because it turns the guest from a lead into a partner in the same week. A warning I learned from experience: not everyone will help you. Some content stars criticized my show after the recording and never shared it. Lloyd shared his episode, and the effect on my own reach was immediate. So I stopped planning for the average guest and started planning for the 10 Lloyds. Assume most won't share. Build for the ones who will. The play. What's in the package? Every guest gets a page, their own page on my site, the video, a transcript, a short summary, key quotes, and the structured data that lets search engines and AI assistants understand it. Guests link to it and it keeps bringing traffic for years. A reshare kit, 3 ready-to-post captions, A, B, and C. And a quote card. Sharing takes one click. A spot in the book. The best quotes from every interview go into one living book with the guest's name, their company, and a link back to them. A tag. When the episode goes live on LinkedIn, an agent tags the guest and their company in the post. They see it, and so does their network. An email that's ready to send. It has the link, the kit, and a line inviting them onto my partners' shows too. Notice what's missing— a pitch. There's no price in the package, no offer, no book a call with my team. The point is to give so much that the next conversation is easy. The play for the guests I most want to build with. Some guests I want to build a real relationship with. The people with large networks and a long runway. I call them GOATs. For them, there's more. 3 short clips cut from their episode, 2 warm introductions to people in my network, a referral to another show where they'd shine, a personal note from me, written by me, not a template, and never a price. I don't offer a low-priced product to a GOAT. They don't want a $100 course. They want a peer. The play: test everything, then keep the winner. The package is also an experiment. I log every guest in a spreadsheet, Which version of the email they got, whether they shared within a week, whether they replied, whether they booked. The first test: when should the GTM Ventures line appear? In the delivery email, A, or only 3 days later, B? Goats always get B. Everyone else is split. I read the result every week and pick a winner. The second test is a paid tier for guests who aren't goats, $100 and $500 versions with extra clips, extra introductions, and extra distribution. I'm not running that until the free version has been through enough guests to see what people actually value. The play? The clips? One episode becomes a month of content. Cut the best moments into short clips for YouTube, LinkedIn, Instagram, and X. Some of mine have reached 10,000 views in 2 hours. Post a lesson from the conversation, tag the guest and their company so they reshare, Write a note to your newsletter about the best insight. Give the guest a kit so sharing takes one click. The reach comes from their audience because they share it. The Play: Books Made Out of Conversations. Interviews are raw material. After a few hundred, I had enough to make books out of them. Founder Wisdom: The Book. The best quotes and stories from every interview in one living book. I add new guests every week. It has an audiobook, read by the AI version of me, and every guest gets listed and linked. The question books. The written answers guests send before their interview, compiled word for word, with their names. The audiobooks. My books have audiobooks in a voice clone of mine, and I say so at the start. This is Charles, the AI version of me. Each book gives me a reason to write to a guest again, and each book gives the guest something to share. A thing they're in. The play. The free book as a follow-up. A week after the call, a guest gets one of my audiobooks. A free book is an easy gift to accept and a good opening to the next conversation. It's the same principle as the invite. Give first. Chapter 12, Step 7, The Close. In October of 2026, I looked at how long my deals actually took, And the number embarrassed me. From the pitch call to a yes or a no, 8 to 14 days, and some were still open. Not weeks of back and forth, days of nothing. One prospect lost 8 days because his co-founder wasn't on the call. In other deals, the proposal followed days after the call, nothing expired, and paying was a separate task from saying yes. Those were my gaps. I had built a system that created thousands of conversations and then left each deal to depend on my memory. So I did to the close what I'd done to every other step. I measured it, found the leaks, and built a small machine. The play, the ask. If you don't ask, the conversion rate is zero. At the end of the interview, after the recording stops, take 5 minutes. Use something specific, they said. Quick question before you go. You mentioned hiring sales reps is killing you. Would you be open to seeing how we'd fix that? That's it. It feels awkward the first time. Do it anyway. And say it in their words. You said hiring is killing you, Lans. We help companies grow doesn't. If a guest is a strong fit, book a 30-minute follow-up before you hang up. A calendar invite made while the energy is high beats let's stay in touch every time. If they're not a fit, ask the other question. Who do you know who'd be a fit for this? A no from the right person is a door to the right person. The Play: 7 Levers That Shorten the Cycle. These are the changes that took my cycle from 8 to 14 days toward a target of under 5. Get the decision maker on the first call. At the pre-interview, ask, who else weighs in on a decision like this? Add them to the invite. Most of my slow deals were waiting for a second person who had never heard the pitch. Send the proposal within 24 hours. Reply in the same thread. Use the numbers they gave you on the call. Never write, let me put something together. Book the decision call before you hang up. A date in the proposal makes the deadline real. Put an expiry on everything. A 72-hour price hold. A limit on start slots, but only a true one. Check your own capacity first. Make paying one click. And give a reason to pay now. Don't chase, nudge on a schedule. If a deal is idle for 2 to 4 days, ask one question that moves it. At five to eight days, nudge with the price hold. At nine to fourteen, say you're closing the file and offer a deposit or financing. After fifteen days, park it for thirty days and move on. Reply today. If the prospect wrote last, that reply is the first thing I do each morning. Every day I wait adds a day to the cycle. The play: the offer ladder. Every proposal has the same four options. I only have to approve the wording once. A, the standard. A 3-month sprint at the standard price. Podcast guests get about 15% off. B, pay in full. Pay by the hold date and start within 48 hours. That start date is the incentive. No extra discount. C, a deposit. 50% now reserves the start slot. The balance is due on day 30. D, financing. The same price, spread over time by a financing partner. And I'm paid upfront. I only offer this when a partner is live. And for guests who think the price is too high, a risk share: 15% off the retainer in return for 15% commission on closed deals. The play, the cycle, nobody counts. The pitch call is only the last part of the clock. A few years ago, I added it up: about 4 weeks for a guest to accept the invite, 2 weeks to get the episode done, And about 6 weeks to a decision. That's 12 weeks from the first email to a signed deal, which is why a feedback loop on a podcast funnel is so slow. It also explains why I tell people to give it a quarter before judging it. The play. Be honest about close rates. A podcast call is not a sales call. A real sales meeting might close at around 30%. A podcast guest closes lower, 10% to 15% in my experience. But the volume is a different universe. At the peak, I had close to 100 podcast meetings a week. 10% of that is a business. And the ones who don't buy right away keep seeing me. The newsletter, the posts, the clips, the retargeting. Many buy months later. The play, the closing machine. Every weekday morning, an agent runs through my calls, my email, and my team chats. It builds a list of every deal that's alive, ranks it by value times probability divided by days left, and writes up to 10 draft replies in my voice. It never sends them. I read them every morning. Some I send as they are, some I rewrite, and some I kill, because the agent can be wrong about a person. That's why it only drafts. The deal room is one spreadsheet. It has every open deal, who has the ball, How long it's been idle, and the next move. 3 things the machine has taught me. Some deals aren't deals. I tell the agent which people are not a fit and it stops proposing them. Some deals need one good question and nothing more. A lot of the delay sat on my side, not the prospect's. If you hate closing, you don't have to. Bring in someone who closes for a living. A great closer turns a lead engine into a cash engine. Give them a share of what they close and give them the context. The interview, the intake, the score. They'll start the call knowing everything you learned. The play, the math, with a warning. The simple version I use to explain it: 100 guests in your exact niche, 5% become clients, so 5 clients at $3,500 a month each. That's $210,000 a year from one campaign. That is an illustration, not a promise. Your niche, price, and close rate will change every line. Run your own numbers before you quote mine. Chapter 13. Step 8: The Loop. Most of my money doesn't come from the first conversation. It comes from the fifth. I've said that for years, and then I went looking for proof. I pulled my 2026 campaign data and compared reply rates. The cold campaigns, the ones going to strangers, replied somewhere between 0.1% and 1%. Depending on how healthy my infrastructure was that week. The warm ones going to people who already knew me replied at 3% to 7%. Warm beat cold by 10 to 30 times. And I had 5,000 or so people who had touched me in some way— guests, replies, applicants, people who said yes and never booked— and nobody was retargeting them on purpose. That's the loop. Everything before this step produces people. This step is what keeps them. Here's what silence looks like. Many CEOs give no feedback at all. And then one day, out of nowhere, one buys and tells me he's been following me for a year. I can't see that coming, so I don't try. I show up every day and let the calendar catch up. The play, the follow-up system. Relationships compound if you keep showing up. The order matters more than the cleverness. Right away, the first hour after the call, thank them and tell them when the episode will be live. Invite them to follow you and say you'll follow back. Ask, any founders you'd refer to the show? 1 to 30 days later, a specific idea for their business based on what they said on the show, a relevant introduction, a case study from a client who looks like them. The episode, the clips, the post, ready to share. 30 to 120+ days later. Something useful now. A list, a tool, an insight. How's business? A simple ROI calculation of what your offer would do for them. My scheduling tool sends most of the plain ones for me. After a call, it fires a series of emails. Within the hour, a note about other shows they'd fit. A few hours later, a request for referrals. A few days later, invitations to my communities. Over the next weeks, the offers. I counted 35 automations before and after around one 15-minute call. That was too many. When I read the replies guests sent, the referral asks were the ones that worked. Any founders you'd refer? Other pods? The confusing duplicates caused questions and at least one cancellation. I'm trimming the ones that get no replies and keeping the ones that get introductions. The play, who to follow up with first. If you have limited time, rank by how warm the person is. People who said yes to an invite and never booked, past guests, people who filled out an intake form, people who were referred by someone, people who replied not now, everyone else. The first 4 are first-party signals. They've already told you something. In my data, they beat public signals by a wide margin. A funding announcement, a job post, a LinkedIn comment. Public signals are worth watching, but they're second. The play. The trigger engine. I have about 19,000 LinkedIn connections. I can't talk to all of them, and I can't read the news for all of them. So I built a small engine that does the reading. It scores every connection on 2 things. How well they fit who I sell to, and how warm they are— calls, emails, bookings. About 300 land in Tier A, 1,400 in Tier B, and the rest are ignored. Every day it scans free news sources for Tier A people, and every week for Tier B. When something happens that signals a budget and a need, a raise, an acquisition, an expansion, a new executive, A new client? It flags them. An AI step then verifies the news is real and about the right company and drafts a note in my voice. 3 rules keep it honest. It checks that the company is the right company. A Rodeo article for a company called Rodeo isn't news about the company. Bad news suppresses. Layoffs or lawsuits mean no outreach. Cooldown. Someone I've nudged waits 60 days. It costs almost nothing to run, because the expensive step only runs when there's a fresh trigger. Stay everywhere. Everyone who's touched me should keep seeing me, until they book or tell me to stop. Not by spam. By showing up in the places they already are. A short newsletter, a few lines, a real story. No headings. YouTube. Retargeting ads to people who watched. LinkedIn. Matched audiences for ads, plus the posts that tag them. Google. Search and display for people on a list. That's the plan, and it's partly live. The inbox and the posts run today. The ad channels are waiting on account setup. I keep this warm only. I never add cold lists to the remarketing engine because the point is to be familiar, not intrusive. The audience at last count? About 5,400 people in segments by what they might want, each with a heat score from 1 to 3. The play? Public intent with a reality check. Some people post online that they're looking for exactly what I sell: a lead gen agency, a cold email firm, help with outbound. I have an agent that searches X, LinkedIn, and Reddit for them every morning and stages replies. The first week told me the truth. It found 185 posts. 12 looked like candidates. 2 were real buyers. Demand like that is sparse, and most of the noise is people hiring setters or selling the same service. Expect between 0 and 3 real buyers a day, and treat it as a cheap trickle, not a source. The play: Measure the leak. Everything above depends on knowing who actually engaged. 3 numbers worth reading every week: Positive replies versus actual bookings. The leak I showed you in step four: reshares within seven days of the package. Replies to the remarketing sequences by segment. If a number goes the wrong way, ask what changed. Chapter fourteen, step nine: The Compound. In early October 2026, I met two people with enormous networks—whales in my head. Not just rich, connected to the exact kind of people I wanted my other guests to meet. The old me would have had a nice call, followed up twice, and moved on. The new me had a problem: I'd already built the machine that finds the right introduction for any person I know. So I wrote down both of them in a registry, ran their networks against mine in both directions, and the machine produced a queue of 19 proposed introductions the same day. Nothing was sent. Every one of those is a draft, and everyone is double opt-in. I ask each side whether they want to meet before I make the introduction. But that was when I saw the compound clearly. One conversation produced 19 proposed introductions. That is step 9. The system pays you back by producing the next input, and the best input is a person. The play. Guests bring guests. Ask every guest, who else should be on the show? A warm introduction books faster than the best cold email ever written. It converts better too. And a guest who recommends someone is putting their name next to yours. Ask twice, right after the call, and again when the episode goes live. Make the second ask specific, like, do you know a founder who just raised? The play, The Golden Habit. One meeting, one referral. No exceptions. Before the conversation ends, I connect every guest to someone useful. In return, I ask for referrals, and I pay 20% commission when one of them turns into a GTM or fundraising client. The ratio is not fair, and I've made my peace with that. I refer them 1 or 2 people a day. They refer me 1 person every couple of months. That's okay. The goodwill bank fills slowly and pays out for years. I also learned what doesn't work. I once recruited salespeople as resellers, and I'd give them my $2,500 a month coaching for free. They delivered nothing. I stopped it, and I kept the 10 to 20 friends who refer me at least one client a month. The play introductions are the product. Become the person who connects people. Every good introduction builds trust with 2 people at once. I run a small system for this. A registry of the people with the biggest networks. A ledger of every introduction I've proposed, so I never suggest the same pair twice. 2 levels deep, no more. The whale's network and my network, and then the whale's referrals. After that, the value falls off. Double opt-in, always. I ask both people first. Drafts, never sends. I read each one and decide. The play: clients become partners. If a client gets a result, they know someone who needs the same thing. Pay them for it. I give partners a structure that grows as they close deals. They start with a small slice of revenue and equity, and the slice climbs with each close. The ladder I'm starting from looks like this: zero closes, 0.1%. 1 close, 0.25%. 3 closes, 0.5%. 5 closes, 1%. 10 closes, 2%. 20 closes, 3%, which is the cap. An agent reads my payments every 2 hours, counts each partner's first paid customer as a close, updates a dashboard, and tells me when someone moves up a tier. It never messages the partner. A person does that. I also keep a leaderboard, fed by payments. Since July first, it shows twelve paid deals. Everyone sees where they stand. A connector in my network, someone who runs a leadership community, referred me a client who signed and another prospect who is in talks. Find the connectors and treat them as a channel. The compound shows up in strange ways. At a Michelin dinner, a client who had become a friend told me a competitor charged forty thousand. a month. And I went into what I call G-force mode. That friend had been a client for about 6 months, left, and became better than me at one of the tools I teach. We're now building a joint venture. The funnel as I draw it, podcast interview, then nurture the 1% performer, then they become a client, then they refer you more clients. Sometimes the student surpasses the master. If you've picked the right people, It's fine. The Play: Communities That Keep Relationships Alive. Deals come months after the episode. In the meantime, you need a place where the relationship stays warm. A weekly meditation sit for founders and investors. A poker night every other Friday for venture investors. Paid hot seat groups where CEOs help each other, which I'm starting with podcast guests. A live pitch show, 5 founders pitching 5 investors in 90 minutes. I run these because I like the people. They also keep the relationships warm between deals. The play: never have a client at zero. The back end of the funnel is the retention ladder. I decided that no client should ever leave me at zero dollars because no is rarely the whole story. Often they can't afford the full thing this quarter and they still believe in it. I wrote the ladder in October and I'm still testing it. It has 13 rungs. The top rung is the full program at $5,000 a month. Then loyalty pricing, Core, Lean, a Performance Hybrid, Done With You, an AI SDR, Software with Strategy, Software Only, an Evergreen plan, and a Keep Alive floor at $500 that covers the infrastructure. Below that, a Hibernate plan at $250 and a Vault plan at $99 that keeps the account and the data safe. Infrastructure moves on to a card owned by my company and licensed to the client, so nothing breaks when the price changes. Renewals follow a schedule: a scorecard at kickoff, a health review at 90 days before the end, a quarterly review at 60, a 3-option proposal at 45, and a call with me at 30. The best time to ask for a renewal is inside 48 hours of a peak, a closed deal, or a 10 out of 10. The play, the onboarding that doesn't depend on me. A new client goes through a 30-day playbook of 29 steps in 7 stages. Each step is either live, scripted, or still being built. A dashboard shows every client, every step, and which ones are blocked. A risk engine flags clients who are stalling so a person can step in. Most of our clients are live inside 10 days. The play. The physical touch. Almost nobody sends mail anymore, so it's remembered for months. A handwritten note, a book with a page marked, a printed copy of the episode transcript with their best line highlighted. It's cheap, and almost nobody does it. Part 3: The Machine. Chapter 15: The Machine— What Runs Without Me? By October of 2026, I had more than 60 scheduled agents running pieces of my business, and my revenue was going down. That's the part I didn't expect. I had an agent for replies, an agent for lead scores, an agent for the daily LinkedIn outreach, one for clips, one for the newsletter, one for the books, one for calendar colors. Every morning, a stack of reports arrived. Then I looked at the money. I make multi-six figures every year, and I'm a multimillionaire at this point. So this wasn't a survival problem. But revenue had slipped from where it was in June, and that told me something. There were a lot of agents and no shared measure of what moved cash. So I added one more agent, and it was the most important. Every morning around seven, it does four things in a row. It measures where we are against the scorecard. It red teams the plan. What's wrong with it? What are we fooling ourselves about? It dispatches the next experiments to the other agents, each with a metric, a target, and a check date. And it learns. It writes down what worked and what didn't. Every experiment goes into a ledger. Nothing gets done twice and nothing gets forgotten. Agents don't fix a business. A scorecard does. Agents just make the scorecard faster. The play, the gates. Agents are fast, so I gave them a short list of rules. They fit on one page. Research, drafts, lists, code, dashboards, docs— just do it. Don't ask. Spend under $25 in a session. Do it. Log it in the report. Anything outbound— email, a campaign, a post, a DM— Stage it. Send nothing. Ask me one line. Send. Spend over $25. Anything destructive. Anything public. Stop that step only. Keep working on the rest. When an agent hits a gate, it finishes everything else first, then asks once with everything I need to say yes. I've changed this over time. For my own campaigns, I've given a standing okay. Launch them. Refill them. Pause the losers within a weekly budget of $150. For clients, every launch still needs my yes. The play. The daily rhythm. Roughly what runs in a day. Rough. Early morning. The business loop, the deliverability guard, the lead scores for the day's calls, the news brief, the search for people asking for help with outbound. Mid-morning. The closing machine, The Did They Book Matcher, the Reply Drafter, the LinkedIn Outreach. Twice a day on weekdays, the Client Sweep, which reads every client's messages, emails, and campaigns and stages what needs attention. Through the day, clips going out, guests getting tagged right after episodes go live, partner payments syncing to the leaderboard, calls getting summarized and filed. Weekly, the Calendly cleanup, the Disc cleanup, and the growth reviews for each show and book. I don't read all of it. An agent whose job is managing the other agents nudges stalled ones, collects what only I can do, and pings me once. The play. The 3 rules that make agents safe. Everything leaves a file. A run that doesn't write down what it did, what it found, and what it needs didn't happen. When something breaks, the files are the only way I can tell what changed. Every routine ends by proposing one new agent— a skill, a helper, or a hook. This is how the fleet grows me planning it. If I'm likely to ask for it again, it becomes a skill. The second time I ask for something, I shouldn't have to explain it. The Play: Where Agents Stop. I learned where agents are not good the hard way. Anything that needs a login. Browsers get signed out, extensions disconnect. A lot of my blocked list is nothing more than a login I haven't renewed. Anything that spends money. I want a human there. Anything permanent— deleting leads, closing accounts, publishing under my name— they prepare it. I click. Taste. An agent can write 100 emails. They can't tell you whether you'd be proud to have sent them. So the real bottleneck is the number of times a day the agents need me. My answer is to make every ask a single click and to put the asks into one short list at the end of the day. In plain words, what I need, Why, and the one button to press. Hire the machine before the people. If you're starting out, build the machine first. Start with one agent that reads yesterday's replies and drafts answers. Review them yourself. Add the guard that watches your domains. Add the one that scores your calls. Then bring a person in and give them the machine to work with. Chapter 16: What Broke? I want to be straight with you about this chapter, because a playbook with no failures in it is a brochure. Everything in this book was paid for. Most of it was paid for in the last few weeks of September and the first weeks of October 2026, when a lot of things broke at the same time. Here's the list, with what each one cost and what I do now. The play, the list. I trusted the warm-up score. It read 100% on every inbox while my replies fell off a cliff. Now I run seed tests to real inboxes and check blocklists every day. I put hundreds of inboxes on one shared hosting plan. The host suspended the account and reset every password. About 850 inboxes went dark overnight. Now, no shared plans for cold email, and any inbox I put back goes in slowly, at low limits. I connected the same domains and inboxes to several workspaces. One burned domain hurt 4 accounts. Now, one domain, one workspace. Volume in one of our workspaces tripled in a day, from 1,900 sends to 5,600, with no limit on it. A guard now pauses any inbox that crosses a bounce or blocklist line, automatically. I let an AI reply to people in my name, unsupervised. It sent the wrong calendar link and confused real prospects. Now, drafts in my voice from the lead's own words, reviewed. Links got into cold emails. I had banned them, and one still showed up in a client's step 2. Now there's a rule and a check that scans every step for a link before a campaign goes live. I launched copy that didn't match the list— business owners to a list of CEOs. Now every campaign gets a read against its own list before it goes out. I measured replies and ignored bookings. 1,068 yeses, 301 calls. I was watching the wrong number. Now a job matches every yes to a calendar slot daily. I couldn't tell where my bookings came from. Now every card shows a source. I kept paying rent on dead lists. 57,000 of 75,000 contacts were in campaigns replying at a fraction of a percent. Now finished leads are deleted nightly after a backup, and the winners are protected. I sent too many automated emails around one call. Duplicate questionnaires, a reminder that mentioned the wrong video tool, a follow-up for a channel I no longer had. Now I audit the sequence and cut anything that gets no replies. I let pitch calls drift. 8 to 14 days from the pitch to a decision, and some still open. Now there's a deal room and a clock. I built on a rented platform. One of my YouTube channels was terminated in October, and the appeal was rejected. I switched off everything built around it. Now, the audience that matters sits in lists I own, not only in feeds. I relied on one place to hold money. A payment account was frozen. Now every proposal goes through one payment account I control. LinkedIn is the place where I've felt that most. In February of 2025, I got thrown off for 48 hours and 1,100 followers disappeared. I went back through LinkedIn's executive escalations. And I wrote that my faith in LinkedIn was restored and that I'd see how long. By April, I'd been suspended six weekends in a row. I've written that they banned my profile twenty times. My answer was to diversify: a newsletter, YouTube, X. It's the oldest lesson on this list and the one I most often forget. The play, the pattern in the list. Read it twice and you'll see it. None of these were clever mistakes. Things nobody was watching: the warm-up score, the volume, the bounces. Things that were automatic and unreviewed: the reply bot, the copy, the links. Things I never measured: bookings, sources, cycle length. Things that sat on someone else's server: the hosting, the channel, the payment account. When something breaks in your system, ask which of those 4 it was. The fix is almost always the same. Someone watches it, someone reviews it, a number measures it, or you own it. Chapter 17, Run It Yourself or Have Us Run It. You now know the whole system, 9 steps, the numbers, and the things that broke. You can build it. Everything in this book is enough to start, and I mean that. A founder who is willing to do the work and who can write a short email and ask a good question can run the smallest version of this alone this month. Or you can have us run it. That's what GTM Ventures is. The honest reason to pick one over the other is how you spend your time. You can spend your next 3 months learning infrastructure, lists, and replies, or you can spend them having conversations with the people those things produce. I'd rather work with founders who play a long game. This system pays off in months, not days, and the people who get the most from it are the ones who treat every conversation as the start of something. The play. The smallest version. If you're alone, with no budget, here's the version I'd run. Write the profile: your best client and the person at their company who'd say yes. Make a list of 200 real people. Warm ones first: past customers, people who replied before, people you've met. Send each a personal note. Not a campaign. One at a time, by hand. I'd love to have you on my show. Mind if I send you more info? Record with whatever you have. A video call. 30 minutes. Make them shine. Ask at the end. Use their words. Send the package within a day. A post they can share, and the recording. Follow up for a year. A calendar reminder per person, every 30 days. That's the whole thing, by hand. You'll learn more from 20 of those conversations than from reading 100 more pages. The play. A 30-day plan. Week 1: Write the guest profile. Build and verify your first list. Set up separate sending domains, mailboxes, and the booking page. Week 2: Launch invites in small campaigns by angle. Answer every reply the same day. Warm up new mailboxes slowly. Week 3: Record your first interviews. Ask at the end of every one. Send the package within 24 hours. Week 4: Start the follow-up system. Cut your first clips. Double down on the angle with the best reply rate. Match yeses to bookings. Most of our clients are live in under 10 days. That includes the part nobody wants to build: the infrastructure. The Play: What we run for you. Your guest profile and the lists to match it. Separate sending domains and warmed mailboxes, watched every day, The invites, written in your voice, tested by Angle, reply handling, booking, reminders, and intake. A score and a source for every call, interview prep for every guest, the package and the follow-up system after every episode, clips and posts from your episodes, the weekly scorecard. Packages and current pricing are on GTM Ventures. The play, if you're already a client. Here's how to get the most out of what we run for you. Show up prepared. Read the intake answers and the guest's last posts before you hit record. Ask at the end of every interview, every single one. Tell us what converts. When a guest becomes a client, tell us who they were and why. We'll book more like them. Share your clips, post them, tag the guest. The reach compounds. Answer the one-click asks. We try to make everything we need from you a single button. When we need a yes, it's because something is waiting on it. Give it time. The best deals often come months after the episode. The play, if you're not a client yet. If you've read this far, you know how it works. The question is whether you want to build it or have it running for you in about 10 days. Reply to the email that brought you here or find us at GTM Ventures. Last page. The whole system fits in one sentence. Invite the people you want as clients onto your show, make them shine, give before you ask, then ask, and follow up for a year. Everything else is execution. Most of it is in the 9 steps, and most of the work is in the boring parts. The list, the inboxes, the numbers. If you run it, tell me what happened. I'll put it in the next edition. If you want us to run it with you, you know where to find me. Charles.