Investment committee mechanics: voting rules that work at scale

The Investment Committee is the single most important governance body in an AIF. It approves every portfolio company investment, every follow-on, and every exit. SEBI does not mandate the IC's composition or voting rules. The fund's documentation governs.

The flexibility creates problems. First-time GPs typically set up the IC at registration without thinking carefully about how it will function across a hundred-plus IC meetings over the fund's life. The structure that works for the first three deals breaks down somewhere between deal eight and deal fifteen, when the team has grown, the schedule has gotten harder to align, and the pipeline has gotten more competitive.

This article walks through IC composition, voting rules, conflict-of-interest protocols, documentation, and the structural choices that work at scale.

Composition: who sits on the IC

The IC has three to five members in most funds we work with. Smaller is brittle (a two-person IC is always unanimous, which is not a real check). Larger is operationally hard to schedule.

The typical 3-person IC. The managing partner, the senior partner, and one external advisor. Works well for a focused fund with one strategy and a single sector emphasis. Decisions are fast. The external advisor provides the independent voice.

The typical 5-person IC. Three internal partners (managing partner, senior partner, head of strategy line), one external advisor, one limited partner observer (non-voting). Works for multi-strategy funds and funds with significant institutional LP base. Decisions are slower but the diversity of inputs is higher.

Avoid the 4-person IC. An even-numbered committee with a tied vote on a 2-2 split is operationally awkward. The tiebreaker mechanism (typically the managing partner has a casting vote) becomes the default decision-making in close cases, which undermines the collective process.

Voting rules that work, ranked

Unanimous. Every IC member must approve. Works for 3-person ICs in highly focused strategies. Breaks down at 5 people because at least one member is regularly travelling or recused, and the practical vote becomes 4-0 with a missing member.

Supermajority (75%). 3 out of 4, or 4 out of 5. Works for 4-5 person ICs. The standard formulation in most LPAs we draft. Allows one dissent without blocking the deal, which preserves the collective process while not requiring perfect alignment.

Majority. 3 out of 5, or 2 out of 3. Used in larger ICs and in funds where dissent is structurally expected. We are cautious about this formulation in smaller ICs because a 3-2 split is too close for the dissenting members to be comfortable; they will eventually leave the IC.

Managing partner override. A clause that lets the managing partner approve an investment that the IC has voted against. We have seen this in first-time fund documents. It is a bad clause. It undermines the IC's purpose and creates a structural conflict between the managing partner and the rest of the team. We recommend removing it during fund formation.

The conflict-of-interest protocol

Every IC member must disclose any conflict of interest before the IC discusses an investment. Conflicts include: personal investment in the target company or a direct competitor, a related party relationship with the founder or management, prior employment at the target, or a personal relationship that affects judgment.

The standard protocol: the conflicted member discloses, the IC chair decides whether the conflict requires recusal, and the recusal is documented in the IC minutes. The conflicted member does not vote and may or may not participate in the discussion (depending on the severity).

First-time GPs often under-document this. The conflict gets disclosed verbally, the recusal happens informally, and the IC minutes do not capture either step. Three years later, an LP audit asks about a specific deal and the GP cannot prove that the conflict process was followed. The fix is administrative: every IC meeting starts with a conflict disclosure round, and the conflicts (or the absence of them) are explicitly recorded.

The IC documentation set

Each IC meeting produces five documents.

Investment memo. Prepared by the deal team. Covers the investment thesis, financial analysis, due diligence findings, valuation, and recommended terms. Standard length 30-60 pages.

Pre-IC question log. IC members submit questions on the memo before the meeting. The deal team prepares responses. This is the most operationally valuable IC document because it forces the substantive disagreement out before the meeting, where it can be analytically resolved, rather than during the meeting where it becomes a personality dynamic.

IC minutes. Decision record, including the vote count, any dissenting opinions, and the conditions (if any) attached to the approval. Standard length 2-4 pages.

Conflict disclosures. As discussed above, captured at the start of each meeting.

Decision register. Cumulative record of all IC decisions across the fund's life. Important for fund administration, LP reporting, and regulatory inquiries.

Common pitfalls in IC design

The 2-person IC. Always unanimous. No real check. The decision is whatever the managing partner says it is. We have seen first-time funds set up with 2-person ICs because the founding team is only two people. The fix is to add an external advisor immediately, even before the first deal is approved.

The 5-person IC where 2 are external. Hard to schedule. The two externals are typically senior people with full-time roles elsewhere, and aligning their calendars with three internal members for two-hour IC meetings every two weeks is an operational headache. The standard workaround is to allow externals to participate by video conference and to vote in writing if they cannot attend. The risk is that the externals become quasi-passive observers rather than active deliberators.

The managing partner override clause. Removes the IC's check. Discussed above.

No quorum requirement. Some LPAs do not specify a minimum number of IC members required for a valid meeting. This means an IC meeting can technically proceed with two of five members present, which is not a real IC. The standard fix: 75% of IC members must be present or attending by video conference for a meeting to be quorate.

No standing meeting cadence. ICs that meet on an ad-hoc basis as deals arise lose the discipline of regular pipeline review. The standard cadence we recommend: bi-weekly IC meetings during the investment period, monthly during harvest. Each meeting has a pipeline review even when no specific deal is on the agenda.

The transition from first-fund IC to multi-fund IC

By the time a GP is raising its second fund, the IC composition needs to be revisited. The first-fund IC was sized for one fund and one strategy. The second fund (especially if it is a different strategy line) often needs a separate IC or a sub-committee structure. We have seen GPs run a single IC across two funds with overlapping mandates and the conflicts and scheduling overhead become unmanageable.

The transition decision: separate IC per fund, or a single IC with sub-committees for each fund. The first works when the strategy lines are clearly differentiated (PE fund vs debt fund). The second works when the funds are vintages of the same strategy.

One committee, many decisions, one purpose

The IC's purpose is to test investment decisions against the fund's strategy and the GP's discipline. Every structural choice in the IC's design — composition, voting rules, conflict protocol, documentation cadence — supports or undermines that purpose. The funds we work with that have well-designed ICs tend to have better track records, not because the IC catches every bad deal but because the discipline of running a real IC keeps the deal team honest about thesis quality and exit visibility. The funds with poorly designed ICs are not always wrong, but they are more likely to be wrong in ways the team did not see coming.