The Keyman Clause: What Happens When Your Fund's Star Partner Walks

TL;DR: A keyman clause (also called a key person provision) is the section of a private fund's limited partnership agreement (LPA) that names the specific individuals whose continued, active involvement the fund is built on, and spells out what happens if they leave, die, get disabled, or stop showing up. When a named partner triggers the clause, roughly 88% of private equity, venture, and credit funds automatically freeze the investment period, meaning the general partner (GP) can't deploy your capital into new deals until the issue is resolved. You need this clause because your commitment was underwritten on a specific team, not a brand name, and without hard trigger language and a real LP vote to reinstate or wind down, you're funding a firm that can quietly become someone else's shop.

Private equity and venture firms sell you a team. The pitch deck has three or four names on it, a track record built by those specific people, and a promise that they'll be the ones sourcing, underwriting, and sitting on boards for the next ten to twelve years. Strip those names out and you're often left with a much weaker case for the commitment you just made. That's the entire logic behind the keyman clause, and it's why the Institutional Limited Partners Association calls the investment team "a critical consideration in making a commitment to a fund" and insists that "any significant change in that team should allow LPs to reconsider," according to ILPA Principles 3.0. If you've written a check into a fund without reading this provision line by line, you've made a bet on people you have no actual leverage to keep.

What the clause actually says, and how it's built

Start with the definitions section, because that's where the fight already happened before you ever saw the document. ILPA's own model LPA defines a "Key Person Event" in almost skeletal terms: it occurs when a named individual "ceases to devote time and attention... as required" or when there's "a Change of Control" at the manager. The named individuals are listed by name in brackets, not by title, not by role, by name. That distinction matters more than anything else in the clause. A provision that protects "the Managing Partners" instead of "Jane Smith and Robert Chen" lets a GP swap out the actual talent and argue the title is still filled.

Once a key person event fires, the mechanical consequence in the overwhelming majority of funds is an automatic suspension of the investment period, the window during which the GP can call capital and make new platform investments. Goodwin Procter's Terms Database for Private Investment Funds, which tracks actual negotiated fund documents rather than model language, found that 88% of private equity, real estate, venture, debt, and infrastructure funds suspend automatically on a key person event, according to Goodwin's analysis of its terms database. That's not a niche protection. It's close to universal, which tells you the market has already decided this risk is real enough that almost no LP will commit without it.

The suspension isn't indefinite. Goodwin's data shows 60% of funds cap the freeze somewhere between three and nine months, and private equity funds specifically lean toward the six-to-nine-month end of that range (40% of PE funds land there), while venture and infrastructure funds track close behind. Real estate funds run shorter, with 55% capping at three to six months, probably because real estate strategies depend less on a handful of named dealmakers and more on operating platforms. Debt funds are the outlier: 30% allow suspensions of twelve months or longer, giving credit managers more runway to sort out succession before the clock forces a decision.

During that suspension window, ILPA's guidance is specific about what the GP cannot do: no recycling capital, no borrowing against uncalled commitments, no new expenditures funded off the balance sheet, unless the LPA expressly allows it. The GP also can't quietly close a deal that was already in the pipeline without looping in the Limited Partner Advisory Committee (LPAC), the small group of LPs, usually five to twelve of the largest commitments, empowered to act on behalf of the full LP base on sensitive matters like this one.

How the suspension gets resolved, and what happens if it isn't

Two paths lead out of a suspension. The GP can propose a replacement key person, which in many funds terminates the suspension automatically once approved. Goodwin found this route resolves the issue in a meaningful share of cases, particularly where the manager negotiated a pre-approved bench. Or the fund goes to a vote. According to Goodwin's follow-up analysis of the same terms database, when a vote is required, about 66% of funds put the decision to a vote of the full limited partner base rather than just the LPAC, and the most common threshold is a simple majority in interest, meaning a majority of committed capital, not a majority of head count, according to Goodwin's analysis of resolution mechanics. ILPA pushes for a higher bar on cause-based events (fraud, gross negligence, bad-faith conduct), recommending a two-thirds supermajority to remove or dissolve the GP entirely, and it explicitly says any such vote should exclude LP interests held by the GP or its affiliates, so the manager can't vote itself out of its own accountability moment.

If nobody resolves the event before the suspension clock runs out, the consequence is severe and, per Goodwin's data, close to automatic: 92% of funds terminate the investment period outright when no resolution is reached. That doesn't dissolve the fund or force a fire sale. The GP keeps managing existing portfolio companies, can still make follow-on investments to protect positions already on the books, and keeps collecting a reduced management fee. What it can't do anymore is call capital for new platform deals. For a fund that's three years into a ten-year life with half its committed capital still uncalled, that's the difference between a normal vintage and a fund that quietly winds down as a shell holding legacy positions, a materially different investment than the one you underwrote.

Real triggers: death, misconduct, and litigated ambiguity

The clause isn't theoretical, and 2023 through 2026 produced several instances that show exactly how it plays out under pressure.

Fortress Investment Group, a $53 billion manager, faced this head-on in October 2025 when co-CEO Josh Pack's unexpected death hit the firm's debt platform. Pack and co-CEO Drew McKnight had run Fortress together for two decades and had just led a Mubadala Investment Co.-backed buyout of the firm in 2023 that put them atop the business, according to Bloomberg's reporting on the leadership shock. Death is the cleanest possible key person trigger. There's no ambiguity about intent or fault, just a binary fact, and it's exactly the scenario ILPA's model language accounts for explicitly alongside disability and retirement.

Odey Asset Management is the messier version, and the one that exposes the real limitation of most keyman clauses: they're built to catch someone's absence, not someone's conduct. When multiple assault and harassment allegations surfaced against founder Crispin Odey in 2023, his partners moved to extract him from the $4.8 billion firm, restructuring two of its oldest funds, OEI Mac and Odey European, to migrate investors into new vehicles under new management, specifically to sever "personal and economic involvement" with him, according to Reuters' reporting on the firm's succession letters to investors. Squire Patton Boggs' analysis of the episode makes the structural point directly: key person clauses "are almost always focused on the devotion of an individual's time... rather than that person's conduct," which means a manager accused of serious misconduct can often cure the technical trigger simply by stepping back or being replaced, while the reputational and redemption damage to the fund has already happened, according to Squire Patton Boggs' review of the Odey situation.

Then there's litigation over what "key man" language actually means when a firm disputes it. In Alphasense, Inc. v. Financial Technology Partners LP, decided by the Appellate Division of the New York Supreme Court in January 2026, the client argued it had validly terminated its engagement because the managing partner had "ceas[ed] his role of actively leading or co-leading the team," missing investor meetings, giving minimal input on fundraising, providing no real guidance. The firm tried to get the case dismissed on the pleadings, arguing the plaintiff hadn't pinned down the exact moment the partner stopped leading. The court refused, holding the key man provision was "open to interpretation" and couldn't be resolved without a full factual record. That's the risk sitting inside vague trigger language. Even a clause that exists on paper can turn into a year-plus court fight over what "actively leading" means in practice.

The most expensive lesson on drafting quality comes from a different angle. A Massachusetts hedge fund partner brought a $630 million legal-malpractice claim against Proskauer Rose after the firm's LPA drafting, described in court filings as a "botched cut-and-paste job," gave the incoming general partner unchecked power to declare a "strategic transaction" and redeem the partner's stake outright, stripping him of an interest worth north of $630 million. A trial judge let the malpractice claim proceed in 2023, finding the drafting failure plausibly caused the loss. Whatever you think about the underlying dispute, the number attached to a single sloppily drafted provision should tell you everything about why you read this section of the LPA yourself instead of trusting that your lawyer's standard markup caught it.

Jeff's take: what you should demand, and what GPs try to slip past you

I've sat across the table from GPs negotiating this clause enough times to know the pattern. They want the key person list as short as possible, sometimes just the named founder, even when three other partners actually source and underwrite the deals. Shorten the list and you shrink the odds any single departure ever triggers anything. Push back on this in diligence. Ask who actually leads deal teams and sits on portfolio boards today, and insist those names go on the list regardless of what the org chart's top box says.

GPs also try to define "time and attention" loosely, something like "reasonable best efforts" instead of a hard percentage. ILPA's guidance calls for key persons to devote "substantially all their business time" to the fund and its parallel vehicles, and that phrase should show up in your LPA in exactly that form or close to it. A soft standard gives a founder room to run a side platform, sit on unrelated boards, or half-commit to a new venture while technically remaining employed at your fund.

Watch the cure mechanics closely. Some LPAs let the GP unilaterally appoint a "pre-approved" replacement with no LP vote at all. That's fine if you actually reviewed and approved that person's name at fund formation, and a rubber stamp if you didn't. Check whether the reinstatement vote excludes GP-affiliated LP interests. If the manager's own commitment counts toward the vote to reinstate itself, you've handed back the exact leverage the clause was supposed to give you.

The single biggest gap, per the Odey case, is that almost no standard keyman clause covers conduct. If your priority list includes reputational risk (sexual misconduct, fraud allegations, regulatory action against a named individual), you need separate "for cause" language layered on top of the standard time-and-attention trigger, with its own lower voting threshold. Do not assume the generic key person clause covers this. It almost never does.

Your keyman clause checklist before you sign a subscription agreement

• Named individuals, not titles: confirm every actual deal lead and board-sitting partner is listed by name, not swept under a "Managing Partners" umbrella.
• Time-and-attention standard set at "substantially all business time," with any carve-outs for side boards or prior fund obligations spelled out and time-boxed.
• Automatic suspension on trigger, not discretionary: confirm the investment period freezes without requiring an LP vote just to start the clock.
• Suspension cap in the three-to-nine-month range. Anything longer than nine months needs a specific business justification from the GP.
• Reinstatement threshold at majority-in-interest of LPs (not just the LPAC) for standard events, and two-thirds supermajority for cause-based removal.
• GP-affiliated interests excluded from any reinstatement or removal vote.
• Separate "for cause" trigger covering fraud, gross negligence, and serious misconduct allegations, independent of the time-and-attention language.
• Clawback test triggered automatically at the key person event, per ILPA guidance, so carried interest already paid out gets tested against fund performance before anything moves forward.
• Clear default consequence if unresolved: automatic termination of the investment period, with follow-on investment rights preserved for existing portfolio companies.

None of this eliminates the risk that the person you backed walks out the door, gets sick, or turns out to be someone you shouldn't have backed in the first place. What a well-drafted keyman clause does is make sure you get a vote, a timeline, and real information before your capital keeps flowing to whoever's left standing.

Author Disclosure: Jeff Barnes, MBA has no personal position in any company, fund, or platform named in this article. Angel Investors Network has no current commercial relationship with any party mentioned. AIN provides marketing and education services, not investment advice. Past performance does not guarantee future results. All investments involve risk, including loss of principal.