AI Prompts for Reading Fund Documents: LPAs, Side Letters & Fee Waterfalls
A distribution waterfall is arithmetic written as prose. The prose is where the ambiguity hides — and where most LPs stop reading.
The five things that matter
- Write the waterfall as a formula, not a summary. Ambiguity survives a summary and dies in a formula. If you can’t write the tier, you haven’t understood the clause.
- The gross-to-net gap widens as returns rise. Carry is charged on profit, so at 1.5x gross the gap is ~0.23x and at 4.0x it’s ~1.08x. One headline scenario tells you almost nothing.
- European vs American isn’t mainly about the total — it’s about timing and clawback risk. On a $50m deal profit exiting six years early, the GP holding $10m of carry that whole time is worth ~$15.9m. ~$5.9m of value transfers even with a perfect clawback.
- The preferred return’s fine print moves the number more than its rate. Compounded or simple, on committed or contributed capital, accruing on fees or not — three details, all easy to miss.
- Let the model read. Never let it calculate. Multi-step arithmetic is where language models are least reliable, and a waterfall is nothing but multi-step arithmetic.
Newly liquid founders committing to their first funds, family office principals reviewing terms directly, and anyone who has signed an LPA they couldn’t fully model.
You have an institutional investment team and counsel already running term analysis — this will be familiar. Also skip if you haven’t yet decided whether the asset class suits you at all.
On this page
Which terms in an LPA actually determine what I receive?
Six. Most first-time LPs read the document front to back and exhaust their attention on boilerplate before reaching the economics. Read these first, then go back for the rest.
| Term | What to find | Why it moves the number |
|---|---|---|
| 1. Waterfall type | European (whole-fund), American (deal-by-deal), or hybrid | Determines when the GP takes carry, which drives time value and clawback exposure |
| 2. Preferred return | Rate · compounded or simple · on committed or contributed capital · accrues on fees? | The fine print moves the figure more than the headline rate does |
| 3. Catch-up | 100% or 80% — and the base it’s calculated on | At 20% carry, a full catch-up equals one quarter of the pref paid |
| 4. Carried interest | Rate · tiered? · on what profit definition | Charged on profit, so its effect grows with performance |
| 5. Management fee | Rate · base (committed vs invested) · step-down date | The base matters more than the rate. 2% on committed ≠ 2% on invested |
| 6. Fee offset | % offset · which fee categories · do credits carry forward? | 100% is now standard; 75–80% remains common. That spread is real money |
If you read nothing else, read the definition of “Contributed Capital” and the definition of “Net Profit.” Waterfalls are built on defined terms, and the definitions are where the economics are actually set.
What is a distribution waterfall, and how do I read one?
A waterfall is the order in which money is paid out. Proceeds flow through tiers, and each tier fills before the next receives anything. The standard four-tier structure:
LP receives =
min(D, C) ← tier 1: capital
+ min(max(D − C, 0), P) ← tier 2: pref
+ 0.80 × max(D − C − P − U, 0) ← tier 4: carry split
where D = total distributions
C = contributed capital
P = accrued preferred return
U = GP catch-up = P × 0.25 (at 20% carry, full catch-up)
Why the catch-up equals a quarter of the pref: a full catch-up runs until the GP holds 20% of profits distributed. Solve U = 0.20 × (P + U) and you get U = P × 0.25. Worth knowing, because it’s the tier LPs most often misread — and it’s the one that looks like a rounding detail and isn’t.
What’s the difference between a European and an American waterfall?
European pays the GP no carry until the whole fund has returned capital plus pref. American allows carry on individual profitable exits before that.
Most explainers stop there and imply the American structure simply pays the GP more. At a fully realised fund with a functioning clawback, that’s not quite right — the totals can be identical. The real difference is timing, and timing is worth money.
A deal exits in year 3: $30m invested → $80m returned. Profit $50m. Fund winds up in year 9.
- American: GP takes 20% of $50m = $10m in year 3
- European: GP takes $0 until the fund has returned capital plus pref
$10m held for 6 years at 8% = $15.87m (1.086 = 1.5869).
So roughly $5.87m of value transfers to the GP even if the clawback works perfectly and every dollar comes back. The clawback returns the nominal amount. It does not return six years of use.
• Is it escrowed? An unescrowed clawback is an unsecured claim against individuals who may have spent the money.
• Is it calculated after tax? Many are — the GP returns only what it kept net of tax paid, so you recover materially less than the gross figure.
• Is it joint and several among the GP principals? If several only, one departed or insolvent principal is your loss, not theirs.
Why is my net return so much lower than the gross?
Because fees reduce the capital actually invested, and carry takes a share of the profit — and the second effect grows as returns rise.
| Gross MOIC | Proceeds | To LP | GP carry | LP net MOIC | Gross–net gap |
|---|---|---|---|---|---|
| 1.0x | $85.0m | $85.0m | $0.00m | 0.85x | 0.15x |
| 1.5x | $127.5m | $127.5m | $0.00m | 1.27x | 0.23x |
| 2.0x | $170.0m | $156.0m | $14.00m | 1.56x | 0.44x |
| 2.5x | $212.5m | $190.0m | $22.50m | 1.90x | 0.60x |
| 3.0x | $255.0m | $224.0m | $31.00m | 2.24x | 0.76x |
| 4.0x | $340.0m | $292.0m | $48.00m | 2.92x | 1.08x |
Two things in that table are worth sitting with.
First: at 1.5x gross the GP earns no carry at all. Proceeds of $127.5m don’t clear capital ($100m) plus pref (~$46.9m). LPs take everything. The preferred return is doing far more work than most first-time LPs realise — and it’s why GPs negotiate hardest over its base and compounding, not its rate.
Second: the gap grows. 0.15x at 1.0x gross; 1.08x at 4.0x. When a GP shows you a 3.0x gross track record, the LP-relevant figure in this structure is 2.24x. Neither number is dishonest. They’re answers to different questions, and only one of them is your question.
Prompt 1: The waterfall translator
The flagship. It converts waterfall prose into an explicit formula — and forces every ambiguity in the drafting to become visible.
## ROLE You are helping me convert fund distribution language into an explicit formula. You are not calculating anything and not advising me. Your job is translation and ambiguity detection. ## THE CLAUSES Distribution / waterfall clause, verbatim: [PASTE — include every sub-clause and cross-reference] Defined terms it relies on, verbatim: [PASTE the definitions of Contributed Capital, Net Profit, Preferred Return, Catch-Up, Carried Interest, and anything else the clause references in capitals. THE DEFINITIONS ARE WHERE THE ECONOMICS LIVE — do not summarise them.] ## PRODUCE 1. THE TIER STRUCTURE. Each tier in order: what triggers it, who receives, how much, and what fills it. Use the document's own defined terms throughout. 2. THE FORMULA. Each tier as an explicit expression with every variable defined. Where the drafting is ambiguous, write BOTH readings as separate formulas and label them Reading A and Reading B. Do not resolve ambiguity by picking the likelier one. 3. THE AMBIGUITY LIST. Every place the language could support more than one calculation. For each: - quote the exact words - state both readings - state which favours the GP - state roughly how much turns on it This section is the point of the exercise. 4. THE DEFINITION TRACE. For each capitalised term used in the waterfall, note whether I supplied its definition. Flag any I have NOT supplied — you cannot translate a clause whose defined terms are missing, and you must not guess at them. 5. NON-STANDARD FEATURES. Anything here that departs from the common structure (return of capital, pref, catch-up, split). Name the departure and who it favours. Do not assume unusual means unfavourable — say which. 6. QUESTIONS FOR COUNSEL. The specific questions a fund formation lawyer should answer, in priority order, phrased so they can be answered without re-reading the whole document. ## HARD RULES - DO NOT CALCULATE ANY NUMBERS. Not one. Produce formulas only. I will build the model myself. - Do not fill gaps from your knowledge of market standard. If something is missing, say it is missing. - Do not tell me whether these terms are good or bad. - Quote the document rather than paraphrasing whenever the exact words matter.
“Do not calculate any numbers. Not one.”
This is the opposite of what people want from AI, and it’s the single most important line in the prompt. Language models are least reliable exactly where waterfalls live — multi-step arithmetic with conditional branches. The model reads and structures. You compute, in a spreadsheet, where you can see every cell.
Prompt 2: Fee archaeology — finding every fee, not the headline
The management fee is the fee you’re quoted. It’s rarely the whole cost.
DOCUMENT SECTIONS: [PASTE: management fee, expenses, fee offset, organisational expenses, broken deal costs, and any clause mentioning fees paid to the GP or affiliates] Build a complete inventory of every economic leakage. 1. THE FEE TABLE. Every fee or expense: name · who pays · who receives · rate/amount · base it's charged on · when it changes over fund life. 2. THE BASE QUESTION. For the management fee specifically: - Committed or invested capital? - Does it step down? When, and to what? - Does the base change after the investment period? Flag explicitly if the fee continues on committed capital after the investment period ends. 3. THE OFFSET ANALYSIS. - What percentage of portfolio-company fees offsets against the management fee? - WHICH fee categories are covered — transaction, monitoring, advisory, directors' fees, break fees? - Are any excluded? Excluded categories are where offsets quietly leak. - Do unused offset credits carry forward, or expire? - Is the offset applied to my share or fund-wide? 4. THE EXPENSE PERIMETER. What can be charged to the fund rather than borne by the GP? Look specifically for: travel, deal sourcing, in-house operating partners, technology, insurance, regulatory and compliance costs. Is there a cap? Note where language is open-ended. 5. THE ACCELERATION QUESTION. Do monitoring agreements permit acceleration of future fees on exit — that is, can the GP collect several years of unearned monitoring fees at once when a portfolio company is sold? If the documents are silent, say so; silence is itself an answer worth raising. 6. WHAT I'D MISS. Based on this document set, which fee or expense mechanisms are common in this fund type but NOT addressed in what I've pasted? List them as questions, not assumptions. RULES: - No arithmetic. Inventory and mechanics only. - Quote clause references so I can find each one. - Do not compare to "market standard" unless I ask — and if I do, mark it clearly as general context rather than an assessment of this fund.
Point 5 is not hypothetical. Accelerated monitoring fees have drawn regulatory scrutiny — the SEC has brought enforcement action over fee-offset practices inconsistent with fund documents. If your LPA is silent on acceleration, that silence is a question for counsel.
Prompt 3: The clawback stress test
A clawback that isn’t escrowed, is calculated after tax, and is several rather than joint is worth a fraction of its headline value.
CLAWBACK / GIVEBACK CLAUSES: [PASTE VERBATIM] ESCROW PROVISIONS: [PASTE, OR "none found"] WATERFALL TYPE: [European / American / hybrid] Answer each, quoting the clause: 1. TRIGGER. What event causes the clawback to be calculated, and when? At final liquidation only, or at interim points too? 2. ESCROW. Is any carry held back? What percentage, held by whom, released when? If nothing is escrowed, state plainly that this is an unsecured claim against individuals. 3. AFTER-TAX OR GROSS. Is the amount returnable net of taxes the GP already paid? If after-tax, note that my recovery is materially less than the headline figure, and that the shortfall is permanent. 4. JOINT AND SEVERAL, OR SEVERAL ONLY. If several only, what happens when one principal has left, is insolvent, or refuses? Who absorbs it? 5. TIME LIMIT. Is there a period after which the obligation expires? Compare that to the fund's expected wind-up, including likely extensions. 6. GUARANTEES. Are there personal guarantees, and from whom? Is the GP entity capitalised enough to satisfy the obligation on its own? 7. THE HONEST SUMMARY. In plain language: if this fund distributes strongly early and performs poorly late, how much of the excess carry would I realistically recover, and how long would it take? RULES: - No numbers. Mechanics and conditions only. - Quote the clause for every answer. - Where the documents are silent, say "silent" — do not infer a market-standard position into a gap.
Prompt 4: Side letters and the MFN election
Other LPs have negotiated terms you may be entitled to elect — if you know the thresholds and the deadline.
MFN CLAUSE: [PASTE VERBATIM] MY SIDE LETTER DRAFT: [PASTE, OR "none yet"] MY COMMITMENT: [AMOUNT] · FUND TARGET: [AMOUNT] Produce: 1. THE TIER MAP. What commitment thresholds unlock which categories of term? Where do I sit, and what would the next tier up require? 2. THE EXCLUSIONS. Which categories are carved out of the MFN entirely — economics, capacity rights, governance seats, co-investment, transfer rights? These carve-outs are where the valuable terms usually sit. 3. THE PROCESS AND THE CLOCK. How and when am I shown other side letters? How long is the election window? What form must election take? Missing the window is the most common way this right is lost. 4. THE ASK LIST. Given my commitment size and this fund type, which terms are realistically negotiable for an investor of my size? Rank by (value to me × likelihood of being granted). Be blunt where something is out of reach — say so rather than padding the list. 5. THE NON-ECONOMIC TERMS. Which rights are worth more than a fee reduction and get asked for less often? Consider: reporting frequency and granularity, advisory committee participation, co-investment rights, excuse/exclusion rights, transfer consent, key person notification, no-fault removal thresholds. 6. WHAT I SHOULD NOT ASK FOR. Requests that would signal inexperience or waste negotiating capital for an investor at my commitment level. RULES: - Do not draft side letter language. That is counsel's work. - Do not assert what is "market" as though it were fact — mark general context as context. - Flag anything with tax consequences for a tax adviser rather than analysing it.
Point 5 is where first-time LPs leave the most on the table. Fee reductions are the obvious ask and the hardest to win. Reporting granularity, co-investment rights and excuse rights are frequently available, cost the GP little, and can matter more over a ten-year fund life.
A real clause, translated
A representative — and deliberately awkwardly drafted — waterfall tier, of the kind that appears constantly.
“Third, 100% to the General Partner until the General Partner has received distributions pursuant to this clause (iii) and clause (iv) below equal to 20% of the sum of all distributions made pursuant to clause (ii) above and this clause (iii).”
Tier structure: this is a 100% GP catch-up, calculated on the pref (clause ii) plus the catch-up itself (clause iii). It is not calculated on clause (iv) distributions, despite (iv) being referenced — the reference is to what the GP has received, not to the base.
Formula:
∴ U = P × 0.25
where P = distributions under clause (ii), the pref
U = catch-up under clause (iii)
Ambiguity flagged — and it’s a real one:
“distributions pursuant to this clause (iii) and clause (iv) below” — on Reading A, the GP’s catch-up target counts only what it receives under (iii). On Reading B, amounts the GP later receives under (iv) count toward satisfying the (iii) target, which would reduce the catch-up. Reading A favours the GP. The circularity is why this drafting recurs — and why it should be tested against a worked numerical example agreed with the GP before signing, not after.
Definition trace: “distributions” is capitalised in the source and its definition was not supplied. Cannot complete the translation without it. Specifically: does it mean gross distributions or net of withholding? That single definition changes the number.
Question for counsel: “Under clause (iii), do amounts received by the GP under clause (iv) count toward the 20% catch-up target? Please confirm with a worked example at a $200m distribution level.”
Notice what the translation produced. Not a number — a question with a worked example attached. That’s the deliverable. Asking a GP to confirm the mechanics with arithmetic at a stated distribution level is entirely normal, costs nothing, and is the single highest-value email you can send before signing.
Level-up: the cross-fund term comparison
This is the part competitors won’t have, because it only works once you’ve done the translation for more than one fund.
Comparing funds on headline terms — “both are 2 and 20” — is close to meaningless. Two funds with identical headline terms can differ by several turns of net multiple on the same gross performance, because the difference lives in the base, the compounding and the offset.
Compare these funds on ECONOMICS, not on headline terms. FUND A TERMS: [PASTE THE SIX TERMS + FEE INVENTORY] FUND B TERMS: [SAME] FUND C TERMS: [SAME] 1. THE NORMALISED TABLE. All six core terms side by side, restated in identical language so differences are visible. Where a fund's drafting doesn't map cleanly onto the others, say so rather than forcing it. 2. THE ECONOMIC DIFFERENCES THAT MATTER. Rank the differences by how much each affects LP outcomes. Ignore cosmetic variation. State direction — which fund's version favours the LP. 3. THE SCENARIO FORMULAS. For each fund, the LP net formula as an expression I can drop into a spreadsheet. FORMULAS ONLY — do not evaluate them. 4. THE HIDDEN DIVERGENCE. Where do these funds differ in ways NOT visible in the headline terms? Look hardest at: fee base, pref compounding, offset percentage and category coverage, expense perimeter, clawback security. 5. THE ONE QUESTION PER FUND. For each, the single question that would most change my view of its terms. 6. WHAT THE TERMS IMPLY ABOUT LEVERAGE. Terms are a negotiation output. Where one fund's terms are notably GP-favourable, what does that suggest about demand for the fund — and where they're LP-favourable, what might that suggest? Frame as hypotheses to test, not conclusions. RULES: - No arithmetic anywhere. Formulas only. - Do not rank the funds overall. Terms are one input among several and I am not asking you to pick. - Flag any term you cannot compare because I have not supplied the definition.
Point 6 is the genuinely advanced move. Fund terms are the residue of a negotiation, so they carry information about bargaining power. Unusually GP-favourable terms in a first-time fund are a different signal from the same terms in an oversubscribed fourth fund. The terms tell you something about the market’s view of the manager, if you read them as evidence rather than as a menu.
Bad prompt vs good prompt
| Weak prompt | Why it fails |
|---|---|
| “Here’s an LPA. What do I get if the fund returns 3x?” | Asks for exactly the arithmetic models are worst at. You’ll get a confident, specific, unverifiable number — and its specificity will feel like accuracy. |
| “Summarise the key terms of this LPA.” | Produces a summary that reads complete and drops the definitions, where the economics actually live. |
| “Are these terms market standard?” | Invites the model to generate benchmarks from memory. It will, and they may be stale or wrong, and you can’t tell which. |
| “Explain the waterfall in simple terms.” | Simplification is the enemy here. The complexity is the content — an ambiguity that survives simplification is one you’ll sign. |
| Strong prompt | Why it works |
|---|---|
| “Do not calculate any numbers. Produce formulas only.” | Puts the model where it’s reliable — structure and language — and keeps you where you’re reliable: a spreadsheet you can audit. |
| “Where drafting is ambiguous, write both readings as separate formulas and label them A and B.” | Stops the model resolving ambiguity silently, which is the failure mode with the largest financial consequence. |
| “Flag any capitalised term whose definition I haven’t supplied. Do not guess at it.” | Defined terms carry the economics. This one instruction prevents the most common serious error in the whole process. |
| “Phrase the output as questions for counsel, in priority order.” | Produces something actionable that respects the boundary between analysis and legal advice. |
What AI must never do here
| Never | Why |
|---|---|
| Calculate your distributions | Multi-step conditional arithmetic is the least reliable thing a language model does, and a waterfall is nothing else. Formulas from the model; numbers from a spreadsheet. |
| Resolve an ambiguity for you | A model asked what a clause means will produce a reading. Ambiguity is a finding, not a problem to be tidied away. |
| Fill in a missing definition | It will confidently substitute a market-standard definition for the one your document actually uses. Those differ, and the difference is the whole point. |
| Tell you what’s market standard | Benchmarks from memory are stale and confident. Cite a current source or don’t use the claim. |
| Replace a fund formation lawyer | These are negotiated contracts with jurisdiction-specific consequences. AI prepares your questions; counsel answers them. |
| Hold confidential documents | LPAs carry confidentiality obligations and may contain MNPI. Enterprise tier with appropriate data terms, decided before you paste anything. |
Which model, and one setup note
Use a large-context reasoning model — LPAs run long and the waterfall depends on definitions scattered across the document, so truncation defeats the exercise entirely. Paste the defined terms alongside the clause; a waterfall clause without its definitions is not translatable, and a model given one anyway will fill the gap rather than refuse.
The verification step that matters more than model choice: build the waterfall in a spreadsheet, then test it at the boundaries — exactly at return of capital, exactly at the pref, and one dollar either side of each tier transition. Tier boundaries are where drafting errors and misreadings surface, and they’re invisible in the middle of a range.
And use a fresh session to check. Models are markedly better at finding errors in material presented as someone else’s than in their own output. Paste the formula into a new session framed as “someone else wrote this — find the errors.”
Market terms, regulatory scrutiny and model capability all change. We re-verify on each review cycle. Confirm anything decision-relevant with counsel regardless.
The Fund Terms Comparison Matrix
All six economic terms across up to five funds, with the waterfall formula pre-built — fill in the terms, read off the outcomes.
◦ Waterfall formula, pre-built
◦ Gross-to-net scenario table
◦ Clawback stress-test checklist
◦ MFN election tracker + deadline
◦ All five prompts, copy-paste
Free. Spreadsheet plus prompts. Unbranded and free to use inside a family office or investment committee.
Upgrade — the interactive version
The Waterfall Calculator. Enter your six terms and it returns your LP net multiple across the full range of gross outcomes, plus the tier-transition points where the economics change — and it shows the arithmetic at every step so you can audit it rather than trust it.
[[LINK WHEN BUILT — showing the working is the differentiator. Every fund calculator online returns a number; none of them let you see which tier your outcome falls in or why.]]
Common questions
What is a distribution waterfall in private equity?
A distribution waterfall is the order in which fund proceeds are paid out. Under a typical structure, limited partners first receive their contributed capital back, then a preferred return, then the general partner receives a catch-up, and everything remaining is split with carried interest going to the general partner. The waterfall is arithmetic written as prose, and rewriting it as a formula is the fastest way to see what you actually receive.
What is the difference between a European and an American waterfall?
A European or whole fund waterfall pays the general partner no carried interest until limited partners have received back all contributed capital plus the preferred return across the entire fund. An American or deal by deal waterfall allows carried interest to be taken on individual profitable exits before the whole fund has returned capital. The difference at a fully realised fund with a working clawback is less about the total and more about timing and clawback risk.
What is a GP catch-up and how does it work?
A catch-up is the tier after the preferred return in which the general partner receives a disproportionate share of distributions until it holds its full carried interest percentage of the profits distributed so far. Under a full catch-up at twenty per cent carry, the catch-up amount equals one quarter of the preferred return paid. A hundred per cent catch-up is common in private equity, while the ILPA model agreement recommends eighty per cent.
What preferred return is standard in private equity?
Eight per cent is the long standing benchmark and is named in the ILPA Principles, with roughly eighty per cent of buyout funds still using it. The rate matters less than three details that are easy to miss: whether it compounds annually, whether it accrues on committed or contributed capital, and whether it accrues on management fees as well as invested capital. Those three change the number materially.
What is a management fee offset?
A fee offset credits fees the general partner receives from portfolio companies, such as transaction, monitoring or advisory fees, against the management fee limited partners pay. Full offset of one hundred per cent has become the market standard, though many funds still operate in the seventy five to eighty per cent range. The offset percentage, which fee categories are covered, and whether unused credits carry forward are all separately negotiable.
Can I use AI to read a limited partnership agreement?
Use it to locate and structure terms, and to generate the questions to put to counsel. Do not use it to compute your economics. Language models remain unreliable at multi step numerical reasoning, and a waterfall is exactly that. The reliable division of labour is that the model extracts the terms and writes the formula, you build the model in a spreadsheet, and a lawyer confirms the interpretation.
What is an MFN clause in a side letter?
A most favoured nation provision lets a limited partner elect terms that the general partner has granted to other investors in their side letters, typically subject to a commitment size threshold, so that larger investors can access a wider set of terms. Certain categories are usually carved out of the election. The practical questions are what the thresholds are, which categories are excluded, and how long the election window stays open.
Why is my net return so much lower than the fund’s gross return?
Because management fees reduce the capital actually invested, and carried interest takes a share of the profit. The gap widens as returns rise, since carried interest is charged on profit rather than on capital. In an illustrative model with a fifteen per cent lifetime fee load, an eight per cent preferred return and twenty per cent carry, a two and a half times gross outcome produces roughly one and nine tenths times net to the limited partner.
Read next
About this guide
Narracomm is a communications and content strategy team. We build and test prompt systems inside live client work and revise them as models and conditions change. [REQUIRED BEFORE PUBLISHING: named reviewer — a fund formation lawyer or experienced institutional LP — with credential and review date shown. This page contains a computed financial model and discusses contract interpretation. Independent review is non-negotiable.]
Freshness
Last updated: 8 August 2026.
Changelog — 8 Aug 2026: first published. Waterfall model built and arithmetically verified — every scenario row balances to total distributions, and GP carry resolves to exactly 20% of profit above the pref. Catch-up identity (U = P × 0.25 at 20% carry) derived and checked. Timing-transfer calculation (1.08⁶ = 1.5869) verified. Market-term references checked against ILPA and current practitioner sources.
Review cycle: every 14 days. Only bump the date when something material changed. Re-dating an unchanged page is explicitly flagged in Google’s helpful-content guidance.
Sources & further reading
- ILPA — Institutional Limited Partners Association — Principles 3.0 and the Model LPA
- American vs European waterfall in private equity
- CSC — Understanding European vs American distribution waterfalls
- iCapital — Understanding private market fund distribution waterfalls
- Alter Domus — How private equity waterfalls work
- Simpson Thacher — SEC charges adviser over fee offset practices
- Winston & Strawn — SEC scrutiny of accelerated monitoring fees
- Management fee offsets — definition and mechanics
- Google — Creating helpful, reliable, people-first content
Scope: general information about fund document mechanics. Not legal, tax or investment advice, and not an interpretation of any specific agreement. All figures are computed illustrations using the stated assumptions — they demonstrate how a structure behaves, and are not benchmarks, projections or representations about any fund. Cash-flow timing alone materially changes outcomes and is simplified here. Fund terms are negotiated and jurisdiction-specific; have any LPA reviewed by a qualified fund formation lawyer before committing capital. Last reviewed: 8 August 2026 · Next review due within 14 days.
