An investment committee is where a fund turns a partner's conviction into the fund's decision. In a small fund, it may be three partners on a call. In a larger one, it may be a formal meeting with voting members and observers. Either way, the committee is the moment when the evidence gathered in screening and diligence is weighed and a decision is made that the fund will live with for a decade.
Knowing how to run an investment committee well is less about the meeting itself and more about what happens before and after it. A good committee sees the same information, forms views independently, argues about the right questions and leaves a record that someone could read in five years and understand. The sections below set out each part.
What the committee is for
The committee has three jobs. It makes the decision: invest, pass, or invest with conditions. It tests the sponsoring partner's case, so that one person's enthusiasm does not commit the fund. And it creates the record of why the fund acted, which matters for learning, for limited partners and for the founder relationship.
Funds use different decision rules, and it helps to know what is common. In the survey of venture capitalists by Gompers, Gornall, Kaplan and Strebulaev, roughly half of funds required a unanimous vote of the partners for an initial investment. A further 7% required unanimity less one, roughly 20% used consensus with some partners holding a veto, and 15% used a majority vote. Smaller funds were more likely to require unanimity. There is no single right answer, but there should be one written answer for your fund.
| Decision rule | Share of funds | Strength | Watch out for |
|---|---|---|---|
| Unanimous vote | About half | Every partner is committed to the deal | One sceptic can block, so dissent may be voiced less openly |
| Unanimous less one | 7% | One strong objection does not block alone | Repeated lone objections can be ignored |
| Consensus with veto | About 20% | Discussion aims at agreement, senior partners can stop a deal | Unclear who holds the veto and when |
| Majority of partners | 15% | Decisions are quicker | A close vote may leave partners uncommitted |
Prepare a packet everyone reads in advance
The committee should decide on evidence it has seen, not on the sponsoring partner's memory of a call. Send a packet at least two working days before the meeting and freeze it: changes after that point go in a short addendum, so everyone knows what changed.
- The investment memo, with the recommendation, the reasons and the main risks.
- The key claims and their status: which are corroborated, which are disputed and which are still unresolved, each with its source.
- Diligence findings and the questions that remain open.
- The proposed terms: round size, instrument, valuation or cap, ownership, any special rights.
- How the deal fits the thesis, including any exceptions and the reason for them.
- Portfolio context: how the deal relates to existing investments, including any potential conflicts.
Research on what investment analyses contain gives a useful checklist. Kaplan and Strömberg studied the internal analyses that 11 venture firms prepared for 67 investments, and found they covered the attractiveness and risks of the business, the management team, the deal terms and the monitoring the firm expected to do. Our guide on how to write an investment memo covers the memo itself.
How to run an investment committee meeting
Most of the value in a committee meeting comes from surfacing disagreement early. A simple agenda helps:
- 1Collect written views first. Before the meeting, each member writes a short view: lean yes or no, the strongest reason, and the biggest risk. Collect them before anyone speaks, so the first opinion voiced does not anchor the rest.
- 2Restate the decision. The chair says what is being decided, under which rule, and whether this is a final decision or a step toward one.
- 3Start with risks. The sponsoring partner spends a few minutes on the case, then the discussion turns to the risks members raised in their written views.
- 4Test the claims. For each risk, ask what evidence supports or contradicts it, and whether the claim behind it is corroborated or only founder reported.
- 5Hear the case against. Ask one member, in rotation, to make the strongest case for passing, whatever their own view.
- 6Decide and state it. Apply the decision rule, record the outcome and any conditions, and confirm who tells the founder and by when.
Keep observers, such as analysts who did the work, in the room to answer questions, but be clear who votes. Short meetings with a good packet usually decide better than long ones that re-pitch the deal.
Record the decision and the reasons
The record is the part most small funds skip, and the part they most often wish they had a few years later. A decision record does not need to be long. It needs to be specific.
- The decision, the date and who took part, with each member's final position.
- The main reasons, in two to five sentences, and the risks the committee accepted.
- Conditions, such as confirmatory diligence items that must clear before closing.
- Dissent, in the dissenter's own words if they want it recorded.
- For a pass, the reason and what would make the fund look again.
- The version of the packet the decision was based on, so later changes do not rewrite history.
Review a sample of past decisions once a year. Compare what the record predicted with what happened, for deals you backed and deals you passed on. The aim is not to score partners. It is to see which kinds of risks the committee tends to over or underweight.
Common committee mistakes
- Deciding in the meeting what should have been read before it. If members are reading the memo for the first time in the room, postpone.
- Letting the most senior voice speak first. Written views before discussion counter this.
- Treating every claim as equally solid. Ask which numbers were checked against documents and which come only from the founder.
- Unclear conditions. "Subject to diligence" means nothing unless the specific items and the owner are named.
- No decision rule. If nobody knows whether one partner can block, the meeting turns into a negotiation.
- Leaving the founder waiting. Tell the founder the outcome promptly, including a clear no. Our guide on how to say no to founders has wording for that message.
Where Delsivo fits
Committee workflow is planned for a later Delsivo release, not the pilot. The design includes investment memos drafted from the deal record for a person to edit, frozen decision packets so later changes appear as addenda, independent reviews collected before discussion, and decisions recorded with reasons and conditions.
What exists in the pilot design is the evidence underneath: a deal record where each claim carries its source and a label of founder reported, corroborated, disputed or unresolved, a thesis screening that shows each criterion with its source, and an audit history of who changed what. AI drafts and cites. People make every investment decision. The claims and evidence page explains the record a committee would read from, and the investment committee software page covers the planned workflow.
Sources
- Gompers, Gornall, Kaplan and Strebulaev: How Do Venture Capitalists Make Decisions? (NBER Working Paper 22587, September 2016) (checked 6 November 2026)
- Kaplan and Strömberg: Characteristics, Contracts, and Actions: Evidence from Venture Capitalist Analyses (NBER Working Paper 8764, February 2002) (checked 6 November 2026)
Delsivo publishes this guide and makes software for venture funds, so we have a stake in the topic. Facts about other companies, research and regulations come from the sources listed, on the dates shown, and can change. Corrections are welcome through our contact page.