Writing
On running the fund.
Writing on fund operations, LP relationships, and the tools that support them.
Architecture
April 18, 2026
The SEC Asked Madoff to Verify Madoff.
A fund's reported figures are only as checkable as the independent evidence behind them. In 1992 the SEC had that evidence one request away, and asked Madoff for a copy instead.
6 min readControl
April 5, 2026
What a Fund's Operations Look Like From the Outside.
Operational due diligence examines how a fund actually operates, then asks the harder question: how much of it can be supported by more than the word of the person who runs it.
6 min readFailure Modes
March 23, 2026
When Is a Mark Wrong, and When Is It Just Different?
A mark that differs from yours could be answering a different question, a defensible difference in judgment, or a mistake. Telling the three apart is the actual work.
6 min readArchitecture
March 16, 2026
A Filing Cabinet With a Search Bar.
Email is structured — around messages. A fund's questions are about investors, terms, and which value currently governs, and none of email's standard message fields holds any of that.
6 min readArchitecture
March 3, 2026
A Paid Distribution Is Not Always Final.
A few provisions in a fund's own documents can require distributed cash to be returned or called again — each narrow, capped, and triggered by something specific.
7 min readFailure Modes
February 24, 2026
Your Administrator's Memory Isn't Yours.
The reasoning behind a fund's own books can live entirely inside one administrator's staff — and an administrator handoff is when the fund learns whether it kept a copy.
7 min readControl
February 11, 2026
Can You Rebuild the Mark You Reported?
A fund can defend a mark only by showing how it was built — work that is necessary to a defense, not proof the number was right.
6 min readFailure Modes
January 29, 2026
A Fund's K-1 Is Only as Timely as Its Slowest Input.
A partnership return can't be finalized until the inputs beneath it are complete — and several of them are outside the fund's control.
6 min readControl
January 22, 2026
The Broken-Deal Expense Is a Fiduciary Question.
A broken-deal expense may be legitimate while its allocation still conflicts with the fund's documents, disclosures, or the adviser's fiduciary duties.
6 min readFailure Modes
January 15, 2026
Two Correct Systems. One Wrong Answer.
Two systems can each execute correctly and still disagree, when only one of them used the term that was actually in force.
5 min readArchitecture
January 8, 2026
Your audit trail knows what happened. It doesn't know why.
Knowing what changed and when is a weaker claim than knowing why.
6 min readControl
December 26, 2025
When does an audit stop being optional?
A first-time manager treats the audit as a cost to defer. An institutional LP treats it as a precondition.
5 min readExecution
December 19, 2025
What really happens when a fund changes administrators?
The risk isn't the decision to switch. It's the parallel-run window where two systems of record both claim to be authoritative.
5 min readFailure Modes
December 12, 2025
When does a waterfall error surface?
The formula was never wrong. The assumption underneath it was.
5 min readArchitecture
December 5, 2025
Why do correct fund numbers still disagree?
Two systems can both be right and still disagree.
5 min readFailure Modes
November 22, 2025
Why don't these fund numbers match?
The LP isn't checking your math to catch a mistake. They're checking it because nobody else is required to.
5 min readControl
November 9, 2025
What does the ILPA Performance Template actually change?
The deadline is 2027. The preparation starts today.
5 min readExecution
October 27, 2025
Why funds keep reconstructing LP conversations.
Every LP promise exists three times. Only one of them gets updated.
6 min readControl
October 14, 2025
Why redemption gates exist.
An evergreen fund promises periodic liquidity, not unlimited liquidity — and the redemption gate is what makes that promise possible.
5 min readArchitecture
October 7, 2025
Why evergreen funds never really close.
A closed-end fund is built around a single close. An evergreen fund is built around a valuation that never stops.
5 min readFailure Modes
September 30, 2025
Why your CRM can't answer the questions that matter.
A CRM's core object is a deal. A private fund's core object is a relationship.
5 min readArchitecture
September 17, 2025
Why later closings cost more.
A catch-up contribution keeps the LPs who funded the fund's early days whole — not a penalty for joining late, but restitution for joining after the risk was already taken.
5 min readControl
September 4, 2025
Why capital call notices needed a common language.
The math behind every capital call was already the same. The language describing it never was.
5 min readArchitecture
August 22, 2025
Capital call notices, explained.
The partnership agreement decides how much an LP owes; the notice just delivers the number.
4 min readArchitecture
August 9, 2025
Most-favored-nation clauses, explained.
Every MFN clause creates a notice process. Most funds still have to assemble it by hand.
7 min readArchitecture
August 2, 2025
Side letters, explained.
Most side letter provisions end with the LP who negotiated them. One doesn't.
6 min readExecution
July 26, 2025
Close readiness isn't built in the last 30 days.
By the time anyone runs the 30-day review, readiness was already decided — weeks or months earlier, by whether the fund's records were kept current.
7 min readFailure Modes
July 19, 2025
The operational debt you can't see.
Memory covers for a fund's missing records just fine — until the people who carry that memory change, and the balance comes due all at once.
7 min readControl
July 6, 2025
How top fund administrators think about document hygiene.
Retrievability was designed in, not hoped for — the difference was never discipline.
8 min readExecution
June 23, 2025
The missing layer in private fund software.
A comparison of DealCloud, Juniper Square, Affinity, and the emerging fund operations category.
8 min readExecution
June 16, 2025
Choosing investor relationship software for an emerging fund.
Every vendor category was built for a different moment in a fund's life. The one that starts after a commitment is signed still doesn't have a clean answer.
7 min readExecution
June 9, 2025
What actually happens after a fund close.
The fundraising literature covers the raise in exhaustive detail and says almost nothing about the years that follow — which is most of the actual job.
4 min readFailure Modes
June 2, 2025
Every fund has a single point of failure.
Institutional knowledge concentrates invisibly in whoever has covered a relationship the longest — and the cost of that concentration only shows up the day they resign.
5 min readFailure Modes
May 26, 2025
Why most private equity funds still run on spreadsheets.
The right tool for eight LPs and one wire deadline becomes the wrong one the moment a second fund turns it into a database nobody designed.
4 min readFailure Modes
May 19, 2025
Your fund already has institutional memory.
Most private funds already have the answers to their hardest LP questions. The real gap is retrieval, not documentation.
6 min read