The Second Moat

Walking in already knowing

Wednesday morning. Marcus called Sarah Kessler (capital markets advisor) from the parking garage at the Driskill at 6:50 AM. She picked up on the second ring.

"Are you here?"

"Two blocks out. What's on your mind?"

"I don't want to surprise you with what I saw this week. Walk in already knowing."

She listened while he walked through Monday morning, the office, the quarterly assembly, Priya's calendar, the IR folder with eight versions of eight quarterly packages each assembled alone. The Tuesday-night unsent draft. The Wednesday-morning question Priya had asked about the IC memo template.

When he stopped, Sarah was quiet for a beat.

"OK. Then we're not spending the morning on positioning. I'm going to walk you through a different conversation. The DDQ's in my bag. Is the corner table free?"

"It is."

"Parking now. Two minutes."

---

The Driskill

Sarah came through the lobby at 6:58 with a leather folio under her arm. They sat at the corner table, the same one where they had worked Fund III's positioning eighteen months earlier. She ordered black coffee and got to it.

"Your returns are fine. Fund II's at 1.7x gross, Fund III's tracking mid-teens net. Fifty-some deals across two chapters of the firm, deal-by-deal years then three funds, and a team that knows what it's doing. None of that's the problem."

Marcus waited.

"The problem's your ops story. You don't actually have one."

He felt the mild irritation of someone being told something he did not agree with. Of course he had an ops story. Twelve years, three funds, competitive returns delivered through disciplined sourcing and hands-on asset management.

Sarah shook her head. "That's a track record. An operating model's a different thing. And the LP sitting across from you in six months isn't the one who backed Fund III."

She pulled out a printed copy of an institutional Due Diligence Questionnaire (DDQ). Twenty-one sections. Over two hundred questions. Operational infrastructure. Reporting capabilities. Technology systems. Data governance. Cybersecurity protocols. Compliance architecture. ESG integration. Business continuity. Fewer than ten questions touched on track record or returns.

"Every question in here's circling one thing," Sarah said. "Can this firm execute at scale without founder heroics?"

Sarah set her coffee down.

"I want to tell you something a senior LP told me in 2022. We were three weeks from closing on his commitment to a fund I was placing. He was the lead anchor. I asked him what made him say yes to this firm and not the seven other mid-market shops he had passed on that year."

Marcus waited.

"He said: 'I'm not betting on the founder. I'm betting on the moat.' Then he asked me, 'Do you understand what I mean by that?'"

"What did you say?"

"I said track record. No. Relationships, proprietary off-market sourcing. No. Operational intensity: boots on the ground, dirty fingernails. No. The GP commitment, the co-invest, the structuring power. No. By then I was grasping. I said brand. He sat back and said, 'You really don't know what I mean. Let me explain.'"

She turned her coffee cup a quarter turn.

What he explained, Sarah said, she had reread from her notes before every placement since. There are two kinds of moats in this business. The first kind lives in people. Track record lives in the founder's pattern recognition. Sourcing lives in relationships the founder has spent twenty years building. Operating fluency lives in the senior asset manager who has run a hundred deals; capital formation lives in the IR head who knows which LP responds to which pitch. All of it real, all of it valuable, and all of it tied to specific people who can leave, get sick, retire, burn out. A moat made of bodies.

The second kind lives in the firm: the structural layer that organizes and amplifies firm intelligence and operating history across time and market cycles. A buy box codified into a screening framework that runs on every inbound. A decision log that captures every IC's reasoning so the next IC's reasoning is better. Operating fluency memorialized into templates, checklists, and harnesses, so the analyst's first IC memo looks like the analyst's hundredth. A capital formation engine that runs whether the founder is on the phone or not. That kind of moat compounds. It survives any specific person walking out the door, and it raises the level every person at the firm operates at.

"Then he said the line I actually wrote down," Sarah said. "'When I commit two hundred million dollars to a fund, I'm betting on the second kind of moat. The first kind is what I'm asking the founder to outgrow.'"

Sarah stopped. Marcus was looking out the window at Sixth Street.

"Marcus, what you're running right now is a firm whose moat is the first kind. Your moat is you, plus Greg's name on the GP entity, Nathan and Priya in the investment seats, Scott and David in asset management, and the sixteen other people who carry the firm out the door in their heads at six and back in at nine. Your returns prove the moat works. From the outside you look like a sports car. The LP who pops the hood is going to find twenty-two talented people pedaling bicycles, keeping the wheels turning one quarter at a time. The motion's coming from the riders, not the car. A sophisticated LP looks at that and sees two dozen points of failure: any one of you has a bad year, or leaves, and the wheels stop."

"You think your edge is sourcing, relationships, your team's ability to grit out a closing weekend. That matters. But it's heroics. The LP across the table in Manhattan reads those as risks. They want to underwrite something else."

"And you've tried to fix it before. The operating system. The president. The consultant you let go for telling you a smaller version of what I'm telling you now. Every one of them dissolved on you, Marcus, and every one was always going to, because each asked you to trust a system before the system had shown you anything. That's an order-of-operations problem. This time the firm's own data does the convincing."

"The second moat. The platform that scales without you, without Greg, without any of the specific people who built it. That's what makes the firm worth anchoring."

She paused.

"Every great CEO I've placed in twenty years built a firm that doesn't need them. Schwarzman built Blackstone to outlast Schwarzman. The firm earns the fifty or hundred million from a family office precisely because it doesn't need you anymore."

"And here's the part nobody tells you. The day the firm stops needing you is the day you've actually built something. An asset. Maybe a legacy."

She let that land, then pushed further. "I want you thinking past this raise, Marcus. The family-office check's the smallest thing on the table. Fifty million, a hundred. That funds the next fund. It doesn't change what you own. Think about how a buyer would price this firm tomorrow. The value lives in your head, mostly, and in the heads of the senior people who run it with you. The day you walk out, most of it walks too. So before anyone opens your returns, they mark the whole thing down: thirty, forty percent, just for that. Owner-dependent firms trade for a fraction of what the same earnings fetch when they don't need the owner.¹ And the carry you're counting on? The promote? A buyer barely credits it. Lumpy, years out, contingent on exits that may or may not come."

"So here's the bigger game. Build the second moat and you're not just de-risking an LP's commitment. You're building a company with value independent of you. One that grows when you're not in the room, that you could scale past your own reach, sell one day, take public, hand to the next generation. You don't have that right now. Right now Chen Capital isn't worth much more than you and the people in it. It's a very good job. It isn't a company yet."

Marcus did not say anything for a long moment. The waiter came and refilled both cups and went away. He picked up his pen.

"Walk me through the DDQ."

Sarah turned to the first page, titled General Information.

"The institutional LP is doing what the family office does in its own way: reading whether the firm can tell its story in two sentences. The raises that move fast share one structural property: the pitch fits in a sentence or two. Basis advantage, yield spread, a margin-of-safety story. The investor gets the edge immediately."

"You'll get one, maybe two sentences with an anchor to describe how this firm actually runs. If the operating model doesn't fit in them, the conversation moves on. That's the test this document administers across two hundred questions."

---

The questions Jordan could not answer

Jordan Wells (Head of Investor Relations) had been fielding the infrastructure questions for six months before Sarah explicitly named them.

The previous quarter, a $200 million pension fund had asked Jordan for a copy of the firm's "technology and data governance policy." Jordan had stared at the email for thirty seconds. There was no such document. She had drafted a two-page summary over a weekend, pulling language from the compliance manual and the vendor contracts. The pension fund's analyst had read it in the first meeting and asked three follow-up questions Jordan could not answer: What is your data retention policy? How do you ensure LP reporting data reconciles to audited financials? What cybersecurity framework do you operate under?

"I made it through the meeting," Jordan had told Marcus afterward. "But I was tap-dancing. They could tell."

The questions had multiplied since. Every new LP inquiry included an operational infrastructure section that was longer than the investment section. Jordan spent more time assembling answers to operational questions than she spent on the actual capital formation conversations she had been hired to lead.

Anika Reeves, the firm's General Counsel, had flagged the governance gaps months earlier. She had walked into Marcus's office one Tuesday with a printed copy of the ILPA's model DDQ and a yellow legal pad.

"I pulled this off ILPA's website and went through it section by section," Anika said. "Two hundred and fifty questions. We can answer maybe sixty with documentation that exists today. Another hundred or so we could cobble together with effort. The rest (I stopped counting at eighty) need policies that don't exist."

She had set the pad on his desk. "Compliance manual needs a full rewrite. The conflict-of-interest policy is three paragraphs from 2018. There's no written cybersecurity plan. The allocation policy across funds is informal. If an ODD team asks how we allocate co-investment opportunities, the honest answer right now is 'Marcus decides.' That doesn't survive institutional diligence."

Marcus had thanked her and put the pad in his drawer. Three months later, it was still there.

(The exercise Anika ran is one any reader can run this week: pull the model DDQ from ilpa.org and count how many of its two hundred and fifty questions your firm can answer with documentation that already exists.)

---

From IRR to DPI

For twenty years, IRR was the number that mattered. IRR rewarded speed: buy fast, add value fast, sell fast. Then the exit market seized. When exits do not happen, IRR becomes a theoretical number. Across the institutional base, DPI (distributions to paid-in capital) has displaced IRR as the first number an LP asks about. DPI does not care about a projected exit timeline. DPI asks: what have you actually returned?

The bar rises hardest for mid-market firms specifically. Large-cap GPs have the infrastructure to manage long hold periods at scale; their operational costs grow roughly in line with the portfolio. At a mid-market firm running on coordination and memory, every extra year of holding compounds the overhead.

McKinsey's 2026 Global Private Markets Report confirmed the structural shift. Fifty-three percent of 300 LPs surveyed ranked a GP's value creation strategy as a top-five criterion in selecting a manager, behind only fund performance and the quality of the investment team. The value creation strategy, the documented, repeatable, demonstrable system for making portfolio companies better after acquisition, is what they are buying.

That is an operating model. That is the second moat.

The implications connect directly to the Coordination Tax and the People Paradox. The team consumed by coordination overhead is the same team that needs to produce the operational narrative LPs are demanding. The DDQ does not fill itself out. The data room does not organize itself. Each capital-formation task competes for time with the coordination overhead that already consumes forty percent of capacity.

This is the bind. LPs evaluate operational infrastructure. Building it requires time and capacity. The Coordination Tax consumes the time and capacity. The loop is closed. And it is compounding.

---

The Compounding Loop

There is a pattern I have seen at mid-market firms that have struggled with fundraising despite competitive returns. I call it the Compounding Loop, and for many firms, it is running in reverse.

When the loop works, strong operational infrastructure produces consistent execution. Consistent execution produces a track record with verifiable data behind it. That track record, backed by the operational narrative, attracts capital on favorable terms because investors can transparently see what they are signing up for. Favorable capital formation funds the next generation of deals with certainty. More deals produce more data. More data trains the systems that drive better execution. Better execution produces better returns. The loop spins faster each cycle.

When the infrastructure is missing, the loop inverts. Capital comes in drips because the operational narrative is murky and fails to engender trust. Dripping capital creates uncertainty at the sourcing level. The analyst screens deals without knowing whether capital will materialize. Uncertain sourcing produces provisional or hedged analysis. Provisional analysis leads to missed deals or compromised deals, where terms get worse during the time it takes to assemble the capital. Compromised execution compresses returns. Compressed returns make the next raise harder. The loop tightens.

---

The unannounced audit

If a diligence team walked into your firm tomorrow on an unannounced audit rather than a scheduled meeting, what would they find?

Would the audit reveal documented workflows and integrated systems producing consistent, auditable data? Decision logs showing who approved what, when, and on what supporting materials? A data room that tells the story of how the firm operates rather than how it wishes it operated? Sourcing logs showing how many deals were screened, where each was rejected, and against what criteria? A quarterly report produced by a system rather than assembled by a person?

Or would they find talented people performing manual workarounds between disconnected vendor systems? A senior analyst who is the only one who knows how the quarterly report gets compiled? A portfolio manager and an asset manager who each assume the other is checking the same assumption? A deal memo that took four people working sequentially because there is no shared context layer? A CFO who is the integration point between accounting systems?

Institutional LPs are pricing operational risk now, in writing, before the returns conversation starts. The diagnostic question underneath every section of the DDQ is the same one: is this firm a system, or a band of heroes?

The loneliest moment in a fundraise is the one that comes quietly. It arrives as a pause rather than a no. It is capital asking you to become something else before it commits to anything new.

---

The second moat

Marcus and Sarah had a working lunch, and he then drove her to Austin-Bergstrom for the boarding call. They sat in his car at the curb of the departures lane while she gathered her bag.

"Here's what I see across my book," she said. "Most institutional LPs run their diligence off a standardized questionnaire now, usually the ILPA template. A fund in market answers a hundred-plus of them per raise. The response window's compressed from two weeks to one. And the ODD section's gone from an afterthought to a gating function. LPs are killing allocations on ops before they ever get to returns."

He nodded. He had been thinking about Priya's quarterly report the entire drive.

"Marcus, your returns get you the meeting. The operating model is what closes the room. Right now you've got a firm that runs on talented people doing manual work. That's a liability with a twelve-year track record attached to it."

She closed the door and walked into the terminal. Marcus started the car, then turned the engine off.

The question that formed was bigger than the Fund IV raise. He was being asked, in effect, to decide what kind of firm he was running. Was this a firm operating as the most rigorous version of what it actually was, or a firm performing the appearance of an institution it had never been?

The Coordination Tax from Chapter 1 was an internal cost. The People Paradox from Chapter 2 was a talent cost. The rising bar was an external force applying pressure to exactly the structural weakness the firm had been living with for a decade. The internal problem had now become an external risk.

His operational model was his pitch. Right now, he did not have one. He had a moat made of bodies. Sarah had named the work for the next eighteen months. Build the second moat.