Chapter 2
Cold Calling Into the Void
The Machine Gets Its Turn
He lasts four days.
Four days of writing emails by hand, ten or twelve a night, each one costing him twenty minutes he doesn't have. Four days of watching the pipeline stay exactly where it was. Eighty-five days left.
On Thursday night Marcus opens the sequence editor again. The 247 contacts are still sitting there, patient as ammunition. He looks at the send button for a long time.
Then he thinks: I'm being precious about this. Ten thoughtful emails got him nothing. Zero is zero whether you agonize over it or not. At least the machine can be wrong at scale — and somewhere in a bigger number there might be a yes.
He clicks.
Then, because the first click is the only hard one, he goes further. He points the enrichment tool at three more lender lists and lets it build him an audience of 2,400. He turns on the follow-up cadence: five touches, four days apart, each one auto-rewritten so no two are identical. He connects the LinkedIn automation. He sets the AI to draft a personalized opening line for every single prospect, pulled from their company page.
It takes him ninety minutes to build a machine that would have taken a five-person team a month.
He goes to bed feeling something he hasn't felt since Rachel's call. Not hope, exactly. Leverage.
Two Weeks of Noise
The numbers come back fast, and at first they look like progress.
By Monday the dashboard is beautiful. Two thousand four hundred sent. Open rate 31%. Click rate 4%. There are little green arrows next to everything. Marcus screenshots it and sends it to David with three fire emojis.
By Wednesday he has replies. Nineteen of them.
Fourteen are unsubscribes. Three are auto-responders. One is a man at a credit union in Tulsa who writes back a single line — how did you get this email — and nothing else.
The nineteenth is from a VP of Lending at a mid-size shop in Phoenix. Marcus opens it with his heart going.
Marcus — your "personalized" opener says you were impressed by our recent expansion into Nevada. We don't operate in Nevada. Whoever sold you this tool should give you your money back.
He reads it four times.
The second week is worse, and it's worse in a way he didn't know to be afraid of.
The follow-up cadence keeps firing. Touch two, touch three, touch four — each one auto-rewritten, each one landing in inboxes belonging to people who already said no. Unsubscribes climb past sixty. Two prospects reply asking to be removed, then reply again, angrier, when touch four arrives anyway because the automation counted their reply as engagement.
On Thursday his email provider flags the domain. Deliverability drops. When Marcus checks, roughly a third of the second week's sends never made it to an inbox at all.
He'd spent two weeks building volume, and the only thing he'd successfully scaled was the reason people wanted him gone.
The phone calls run in parallel, and those he still has to make himself.
Forty-seven today. Forty-seven times he dials, rehearses the opening line, waits. Voicemail. Voicemail. "He's in a meeting." Voicemail. "Can you send an email?" Voicemail.
Three people answer. Three actual human beings who pick up and hear Marcus say, "Hi, this is Marcus Reyes from VaultPath. Do you have a moment to talk about client retention?"
The first says, "Not interested," and hangs up.
The second listens for thirty seconds, asks Marcus to send some information, and has since ignored two follow-ups — both of which, Marcus realizes later, were sent by the machine.
The third is the worst.
"Look," the man says, cutting him off mid-sentence. "I get five of these a week. Five. And lately they're all the same — same structure, same fake compliment about something we did, same fifteen minutes. You can always tell. Unless you can say something specific about my business, something that shows you did actual work instead of pointing software at me, we're done here."
Click.
Marcus sits in silence for a full minute. Then he dials the next number, because that's what the course said to do. Push through. The next call could be the one.
It isn't.
David finds him at 9:47 PM, still at the office, the glow of the dashboard reflecting off his glasses.
"You're still here?"
Marcus doesn't look up. "I sent twenty-four hundred emails in two weeks. Twenty-four hundred. You know how many real conversations that produced? Zero. I made forty-seven calls today and had three, and all three told me to go away."
David pulls up a chair. "Maybe the product—"
"It's not the product." Marcus finally turns. "They don't even let me get to the product. I'm just noise to them. I'm the guy they avoid. I'm the number they don't answer."
He looks back at the screen. Two weeks of effort, catalogued in rows and columns and cheerful green arrows. A monument to failure with excellent analytics.
"And the machine didn't fix it," he says, mostly to himself. "It just made me louder."
The Emails He Would Delete
Marcus drives home in silence. No podcast. No music. Just the hum of the highway and the weight of the numbers.
Somewhere around the Mopac exit, it hits him.
He thinks about his own inbox. The emails he deletes without reading. The calls from unknown numbers he sends to voicemail. The LinkedIn messages he ignores because they all open with a compliment about something he did and close with a request for fifteen minutes.
He hates those emails. Everyone hates those emails.
And for two weeks, at industrial volume, he has been sending them.
That's the part that sits wrong. Not that he sent bad emails — he'd suspected that already, back when he was writing them one at a time. It's that he handed the bad email to a machine and the machine did exactly what he asked. Perfectly. Twenty-four hundred times. It never got tired, never hedged, never had the decency to hesitate over the send button the way he did.
The gurus told him this was how it worked. Outbound at scale. It's a numbers game. Make enough calls, send enough emails, and eventually the math works in your favor.
The math didn't work. And Marcus is starting to suspect it's not because the numbers were too small.
He'd assumed the machine was the missing piece. What the machine actually did was answer a question he hadn't thought to ask: what happens if I do this wrong, but ten times faster?
Now he knows.
The Buyer Who Doesn't Need You
Marcus isn't failing because he's bad at sales. He's failing because he's selling into a world that stopped existing, using a tool that made the old approach worse instead of better.
Twenty years ago, salespeople controlled information. If a mortgage lender wanted to learn about retention technology, they had to talk to a salesperson. There was no Google, no G2, no Reddit thread, no LinkedIn discussion. The salesperson was the gatekeeper to knowledge, and the gate was the leverage.
That leverage is gone. Buyers now complete the majority of their research before they ever contact a vendor.¹ By the time someone takes your call, they've narrowed the field. They may have already decided.
That much was true before AI. Here's what AI changed.
The trust deficit, now with a shorter fuse.
When you cold call someone, you're asking for their most valuable resource — time — on the strength of nothing. No relationship. No credibility. No reason to believe you're worth the minutes. Interruption breeds suspicion, and you begin the relationship in a hole you have to climb out of.
What's new is how fast buyers now reach for the shovel to bury you. They have been trained, over the past few years, by a flood of machine-written outreach that is grammatically perfect and completely empty. They have learned the tells: the fake-specific opener, the compliment that doesn't quite fit, the tidy three-sentence structure. The man who hung up on Marcus wasn't reacting to a cold call. He was reacting to a pattern.
Every generic message you send now does double damage. It fails, and it reinforces the filter that will catch your next one.
Rent versus own.
Here is the thing Marcus hasn't yet said out loud, though he's one day from saying it across a coffee shop table.
The tools he bought are available to everyone. The sequencer, the enrichment, the auto-personalization, the model writing the opening lines — his competitors can have all of it by the end of the afternoon, for about what he paid. There's no version of this where he out-subscribes anyone.
Anything you rent, your competitor can rent too. Which means whatever you rent cannot be your advantage.
That sounds obvious written down. It is not obvious in practice, because renting feels like progress. Marcus's dashboard felt like progress. The green arrows were real. Volume went up thirty-fold and the only thing that changed was the speed at which people learned to ignore him.
The tools are leverage. Leverage multiplies whatever you already have. Multiply nothing and you get nothing — just faster, and with better reporting.
What buyers actually want.
Here's the uncomfortable truth: buyers don't want to talk to salespeople. They want to talk to experts.
If you had a complicated tax question, would you rather talk to a random accountant who cold-called you, or the CPA who wrote the article you found when you went looking for answers? If you needed surgery, would you pick the doctor who mailed you a flyer, or the specialist whose talk you sat through at a conference?
Expertise creates trust. Trust creates access. Access creates opportunity.
People avoid salespeople but pay thousands to talk to experts.
Read that again. The same prospect who screens your call and deletes your email will happily spend five hundred dollars to attend a conference where an industry expert is speaking. They'll read three thousand words from someone who genuinely understands their problem. They'll argue in the comments of a post that names a challenge they're living with.
The difference isn't the message. It's the messenger.
When you cold call, you're a salesperson trying to get something. When you share real insight, you're a resource offering something. Different position, different result.
And notice what that means in an era of infinite generated text: the scarce thing is no longer content. Content is free now, and worth roughly what it costs. The scarce thing is someone who actually knows.
A different question.
What if Marcus doesn't need to get better at cold calling? What if he doesn't need better scripts, better tools, or bigger numbers?
What if instead he could become the person prospects seek out — the recognized expert, the trusted voice, the name someone mentions when a colleague asks, "Who should I talk to about client retention?"
That's not traditional selling. It's something else.
And it starts with a question Marcus hasn't asked: What do I know that other people need to learn?
The Post That Stopped His Scroll
It's 10:47 PM. Marcus is in bed, doom-scrolling LinkedIn. Sarah is asleep beside him and he should be too, but his brain won't stop running the numbers. Twenty-four hundred emails. Forty-seven calls. Zero.
He scrolls past the usual. Motivational quotes. Humble brags. "I'm thrilled to announce" posts that make him feel worse about his own trajectory.
And a great deal of something else, which he's only now learning to recognize: posts that are smooth and confident and say nothing. Seven Lessons I Learned. The Future of Lending Is Here. Paragraphs with the cadence of insight and none of the substance, arriving at a rate no human could sustain. His feed has the same quality his outbox had. Enormous volume, perfectly formatted, weightless.
Then something catches his eye, and what catches it is friction.
The post is from someone named Jennifer Huang. Her title says VP of Operations, Summit Lending. Not a vendor. Not a consultant. An actual practitioner.
"Unpopular opinion: most mortgage CRMs fail because they treat client retention as a marketing problem instead of a relationship problem. Your past borrowers don't need more emails. They need to feel like their loan officer remembers them as a person, not a transaction. The tech should enable that relationship, not replace it."
Two hundred and thirty-seven likes. Sixty-two comments. People engaging, disagreeing, telling their own stories.
Marcus reads it twice, and the second time he notices why it stopped him. It isn't well written, exactly. It's a little blunt. It picks a fight. It says a specific thing that a specific kind of person could only believe after watching it happen — and it is therefore the one thing in his entire feed that could not have been generated by pointing software at a topic.
No machine produces "unpopular opinion" and means it. A machine has no opinions to be unpopular. It has an average, and the average is never unpopular, which is exactly why it never cuts through.
He reads the comments. Loan officers venting about systems they hate. Operations leaders sharing what's worked. Someone asks Jennifer what technology she'd recommend. She answers thoughtfully and names three options, one of which competes with her own vendor. No pitch. Just knowledge.
He clicks her profile. Her last twenty posts are all like this. Specific observations. Problems articulated so precisely that people feel seen.
She isn't selling. She's just telling people what she knows. And hundreds of people in his industry are paying attention.
Marcus screenshots the post. Opens his notes app. Types:
"She never sent me anything. I found her."
He looks at that sentence for a while.
Two weeks ago he'd have read this post and thought: I should post more. Tonight, with a flagged domain and sixty unsubscribes and a man in Phoenix who told him to get his money back, he understands that posting more is the same mistake in a new outfit. Volume is not the variable. Jennifer Huang doesn't have reach. She has something to say.
His phone buzzes. A text from Rachel.
"Checking in. How are the sales calls going?"
He stares at it for a long moment. Thinks about the dashboard, the green arrows, the man who said he was just noise.
He doesn't answer.
But as he sets the phone down, something shifts. The exhaustion is still there, and the seventy-one days, and the sick feeling about the domain. Underneath it, though, is something he hasn't had since this started.
Not a plan. A direction.
Jennifer Huang isn't making cold calls. She isn't grinding a sequence or working a numbers game. And people are coming to her — asking her opinion, trusting her read.
She knows something he doesn't. Not a tactic. Something underneath the tactics.
He falls asleep with the question still running.
Tomorrow he's going to go find the person who can explain it.
Rachel's introduction had mentioned a name. A former founder who'd "figured out a different way to sell."
Tom Bradley.
Endnotes
- 6sense, "The B2B Buyer Experience Report 2024" (survey of 2,509 B2B buyers globally, 49% VP-level or above), https://6sense.com/science-of-b2b/2024-buyer-experience-report/. Buyers report completing the large majority of their evaluation before contacting a vendor; corroborated by CSO Insights/Miller Heiman Group, "Buyer Preferences Study 2018" and Challenger Inc. longitudinal research (2008–2024). See also Chapter 1, note 3.
Draft completed: December 2025 | AI-era rewrite: July 2026