Chapter 9
The Enterprise Slog
The Room That Had Never Heard of Him
The ballroom in Fort Worth holds about two hundred people, and a year ago not one of them would have taken his call.
They are here because the Texas Mortgage Bankers Association put his name on the program, wedged between a compliance update and the lunch buffet, and because enough of them had read something he'd written to fill the chairs. He has forty-five minutes. He spends the first of them certain he is about to be exposed — that someone will stand up and point out he is a software founder who a year and a half ago couldn't close a cold call, let alone a room.
Then he stops trying to be a speaker and does the only thing he is actually good at: he talks about retention the way he'd talk to Tom over coffee. No pitch. No slide worth remembering. Just the real mechanics of why borrowers leave and what every person in this room could see about it in their own data if they went looking. The heads come up. The phones go down. When he opens it to questions the questions don't stop, and a man from a lender in Amarillo says, half to the room, "Why has nobody put it that way before," and the honest answer — which Marcus keeps to himself — is that most people trying to sell you something have never been curious about it for its own sake a day in their lives.
Afterward, the thing that never used to happen happens. A line. Not a long one; this is a regional conference, not a stadium. But six or seven people wait to talk to him, and two of them describe problems specific enough that they are, he realizes, already half a deal. He gives them his full attention and not one word of a pitch, takes their cards, and drives back having done nothing that resembles selling and more of it than a month of cold calls ever produced.
It is the flywheel he had only ever theorized, made suddenly, physically real: the expertise draws the audience, the audience fills the room, the room hands you the pipeline. A year ago he fired twenty-four hundred emails into a void and it answered with how did you get this email. This week a void's worth of people drove to a hotel ballroom to hear him, and he never had to send a thing.
He is still turning that over when the enterprise deal reminds him it runs on a different clock entirely.
The Circle-Back
The call starts at 2:00 exactly, which Marcus has learned to read as its own piece of information. Mike Donovan is SVP of Sales Operations at First National — a top-ten lender, the largest company that has ever written to Marcus about anything — and Mike does not appear a minute early or a minute late.
"Marcus. Thanks for the assessment." Mike's video square is a corner office with the blinds half-drawn. "My team ran it. Came out at eleven out of twenty."
"That sounds about right for where you are."
"It sounds low."
"It's honest. If it came back nineteen you'd have a reason not to trust it — and so would I."
A pause. Mike almost smiles, and decides not to.
"Here's my problem with you. Not you specifically." He sets down whatever he's been holding. "We've had three failed implementations in the last five years. Every one of them was sold to me as the answer. All three were sold to me as AI-powered." He says the phrase the way you'd hold up something you found under the sink. "The first one automated our birthday cards. Very intelligently. Cost us six months and a change-management consultant. So tell me what makes you different from a company that takes nine months of my people's time and leaves us with smarter birthday cards."
Marcus has a sentence ready — our platform is different — and lets it die, because it is exactly the sentence the last three vendors led with.
"I don't think your problem is that your follow-up isn't automated," he says. "I think your problem is that four hundred loan officers can't see their own retention numbers, so retention isn't anybody's job until a borrower's already gone. No amount of AI fixes a thing nobody can see. First you make it visible, then you make it somebody's job, and only then does automating any of it help. If a vendor started with the automation, I understand why it didn't take."
Mike is quiet for a moment. "That's roughly what my head of ops said. Except it took her two years and three vendors to say it."
"She's right. I'd start where she is."
"Send me nothing for now," Mike says. "Let me socialize this internally. These things take the time they take." He looks at something off-camera. "I'll be in touch."
The call ends at 2:41.
Then nothing.
A week. Marcus sends one short note — a servicing-transfer analysis he wrote that happens to bear on something Mike mentioned — and gets back a single line: Useful, thanks. Two weeks. Silence. Three. Marcus catches himself refreshing his inbox the way you'd worry a loose tooth, and understands that he is doing the thing he tells other people not to do. His whole body is on a ninety-day clock, and Mike is on no clock Marcus can see, and the mismatch is starting to feel like a verdict.
He calls Tom.
What Tom Knows About Oak Trees
"How long," Marcus asks, "does a company like First National take to say yes to something like this?"
"Nine months. Could be twelve."
Marcus's voice does something undignified. "I've been counting in weeks."
Tom laughs, not unkindly. "Welcome to enterprise. Here's the part you need to hear: Mike isn't ghosting you. He's following his process, and his process is built to prevent the exact three disasters he described to you. The slowness isn't him being difficult. It's the scar tissue. You're not fighting Mike's caution — you're standing behind the last three vendors who earned it, and honestly, they're helping you. They taught him to go slow, and slow is the only speed at which a careful buyer ever says yes."
"So I just wait."
"No. Waiting is what kills these. You play the long game, but you play it on purpose." A pause. "Think about the difference between an oak and a tomato. The tomato is fast — plant it, feed it, eat it, do it again next season. Quick money, but you're always replanting. The oak takes years. But once it's in, it's in. Nobody uproots an oak because a cheaper oak showed up."
"I need both."
"You need both. And here's the part that's going to sting." Tom's tone changes. "You told me you've got about three weeks left on Rachel's clock. Mike is a nine-month deal. So Mike cannot be your plan. Mike is a tree you're planting in a yard you might not own by the end of the month. Plant him anyway — plant him carefully. But you pay this month's rent with something that closes this month. If you let the oak convince you to stop growing tomatoes, you'll starve in the shade of a tree that isn't done yet."
Marcus writes tomatoes on a Post-it, then remembers the weekend he spent burning his Post-its, and types it instead.
Two Clocks, and One Cadence
So the enterprise deal gets planted, and the near-term deals get harvested. What you cannot let the big deal do is make a near-term win feel small — and it will try. A deal the size of First National has a gravity that pulls your hours toward it and away from the three unglamorous opportunities actually keeping you alive. Name that gravity so you can resist it, and be honest about which clock is paying you this month.
Then, for the deal you've planted, the real work: staying present for the long haul without becoming what Mike deletes on sight.
Because the old enterprise play is dead, and AI is what killed it. The old play was stay top of mind — check in, circle back, touch base, keep your name in front of the buyer. That advice was written for a world where a check-in was scarce and therefore meant something. In a world where a thousand automated sequences check in with Mike every week — each one warm, personalized, and empty — top of mind is worthless. Everyone is top of mind. Which is the same as no one.
The bar moved. A touch now has to survive a single test: would this have been worth Mike's time even if he never buys from me? An insight from a domain you actually understand. A person worth meeting. A way of thinking about his problem he didn't have yesterday. If it doesn't clear that bar, it isn't a touch — it's noise wearing your name, and every piece of it confirms Mike's fatigue instead of fighting it. You are not staying top of mind. You are trying to be the one useful signal in a channel that has become pure static.
Here is where the machine earns its place, and where it doesn't.
The machine remembers the cadence — that it's been nineteen days, that the last thing you sent was the servicing-transfer piece, that Mike opened it twice and forwarded it once to someone whose name you now have. It drafts the connective tissue. It surfaces which of the forty things you've written this quarter actually bears on the eleven-out-of-twenty Mike scored. It does the remembering that used to take an SDR and a coordinator, which is the only reason one person can keep twenty enterprise-grade relationships warm at all.
What it cannot do is judgment. It cannot know that a compliance officer at a thrice-burned lender needs to hear about failure modes before features. It cannot feel that this particular week is quarter-end and any touch at all would land as an interruption. It cannot decide that the right move, this month, is to send nothing and let the silence say you respect his process. The cadence is machine-run. Every decision inside it is yours. Get that division right and you scale attentiveness. Get it backwards and you've built a very efficient machine for becoming static.
The deal is not a person. An enterprise sale is a building full of people who each want something different, and the mistake is selling to whoever answers your email. There is an economic buyer who owns the budget and fears the risk. A technical buyer who will judge whether it breaks anything. The people who'll actually use it, who care about their Tuesday, not your ROI slide. And, if you're fortunate, a coach on the inside who wants you to win and will tell you what the others won't say on a call. The machine can track who has seen what and who has gone quiet. Only you can decide what each of them actually needs to hear, and in what order, to move a decision that no single one of them can make alone.
And the slog is the moat. The thing that makes enterprise unbearable — seven months of near-silence, no way to force it — is the exact thing protecting you. A bot cannot wait seven months. An automated sequence gives up, or gets caught, or gets unsubscribed. The one seller still there in month seven — still useful, never desperate, having quietly built understanding while everyone else churned through their outreach quotas — is doing the single thing the flood structurally cannot do. Mike's caution isn't the obstacle. It's the filter. And it filters out everyone who isn't willing to be patient and expert for longer than a machine can fake it.
The Note to November
Over the long courtship of First National, Marcus sends Mike exactly four things.
A short analysis of how top-ten lenders were handling servicing transfers, because Mike had used the phrase in passing and Marcus knew something about it worth knowing. An introduction to a VP at a non-competing lender who'd solved the visibility problem the hard way — Marcus made the introduction and then got out of the thread, which is the whole art of it. A framework, vendor-neutral, for evaluating a retention platform without getting sold one; it happened to ask the four questions First National would have to answer no matter whose software they bought. And once, when it was quarter-end and any message would have landed as an interruption, nothing at all — the touch was the restraint.
Each one, he decides himself. Each one, the machine remembers, times, and drafts around, so that the deciding is all Marcus has to spend.
The reply that matters comes on a Thursday.
It is not thanks, still evaluating. It is three sentences, and the middle one is a question — a specific one, about how the visibility layer would handle branches that run on a different core system — the kind of question a man asks only when he has stopped evaluating whether to move and started imagining what moving would look like. At the bottom: Copying Dana, our head of ops. She'll have opinions.
A second name. A stakeholder, surfacing on her own.
It is not a close. Marcus makes himself write that down too, because the same instinct that reads three weeks of silence as death will read one good email as a signed contract, and both are the same error with better manners. Mike is still, by Tom's reckoning, months from deciding. This is week eleven.
But it is the first evidence the tree took root.
He opens Mike's record. It's all there — the assessment, the eleven, the four things he sent and the one he pointedly didn't. And near the top, in a column no software ships with, a number and two sentences in his own handwriting from a weekend back in the spring: First National — 58. Conviction: high. He wrote four paragraphs about his own problem to a stranger, and I have no field for that. Check in November whether I was right.
He adds one line under it — the date of the Thursday email, and root take — and closes the record.
Then he looks at the rest of his pipeline, the part that is actually paying him. The Colorado lender is at proposal. Karen Walsh is one conversation from a decision he can almost taste. Two of the mid-market conversations on the board are men who stood in that short line in Fort Worth to shake his hand — the first pipeline he has ever pulled off a stage instead of a phone. The tomatoes are coming in. The oak is planted and, for the first time, visibly alive.
Barely three weeks on his own clock; months, still, on Mike's. Marcus has finally stopped trying to make them agree, and started running both.
The slog, he's decided, isn't the tax you pay for the enterprise deal. It's the price of admission that keeps the machines out — and the one bill a patient expert can afford that a bot never can.
Endnotes
The figure in the first draft — a precise median for the enterprise software sales cycle — is gone. Real benchmarking exists, and it broadly supports what Tom says: deals of this size and risk take many months, often the better part of a year, because they run on committees, compliance reviews, and procurement, not on a seller's urgency. But the specific numbers in circulation vary wildly by how each study defines "enterprise," "cycle," and "close," and I could not pin a single median to a defined population and method I'd be willing to print as fact. So the claim lives where it belongs: in Tom's mouth, as a working operator's experience — "nine months, could be twelve" — attributed to a career, not to a study that turns out to be quoting another study.
Mike's assessment score, the "58" and the conviction note carried over from the previous chapter, and the four-touch sequence are invented particulars. The pattern underneath them is not: in the AI era, a follow-up that carries no real value is worse than no follow-up at all, because it costs you the one distinction that still closes enterprise deals — being the signal and not the noise.