Investment Committee Memo Template

The economists Steven Kaplan and Per Strömberg studied the memos behind 67 investments by 11 venture capital firms¹, made between 1987 and 1999. The memos weighed the business, the management and the deal terms. Every one gave reasons to invest, and all but two named risks. Some were two-page write-ups; others ran past 20 pages.

If your firm has no house format, use this order, with the decision first.

Build a sheet before you draft a sentence. Give each headline figure a row: the number, the document and page it came from, its definition in words, and a status. Stated means a document gives the figure. Derived means you calculated it and show the inputs. Absent means nothing you have supports it.

Two honest figures for the same metric can differ by definition alone. In the US, SEC guidance expects a public company that reports a key metric to give a clear definition of the metric and how it is calculated². When the method changes, the company should consider recasting earlier periods. A private company's deck is not bound by that guidance, so ask for each definition and any restated history.

Where a figure is absent, write absent and request the document. Never fill the gap with a benchmark or a number from an old deck. If you only have the deck so far, a pitch deck analysis lists its figures and the questions to ask first.

A list of worries tells the committee little. For each risk, write the evidence for it, the evidence against it, the number it would move and the document that would settle it. Record what has already happened as a fact: accounts already on month-to-month terms are not a forecast.

Expect the same factor to show up on both sides. In Kaplan and Strömberg's sample, investors saw some aspect of management as a risk in 61% of the memos¹. That is about the same share, 60%, that named the quality of management as a reason to invest.

Before you finalize the list, run a premortem: assume it is three years later and the investment has failed, and have each person on the deal write down why. In Harvard Business Review, Gary Klein cites 1989 research finding that imagining an event has already happened increases the ability to correctly identify reasons for future outcomes by 30%³.

Venture investors most often measure a deal by the multiple of their money it could return. Paul Gompers and colleagues surveyed 885 venture capitalists in 2015 and 2016: 63% used cash-on-cash multiples and 42% used internal rate of return⁴, while only 22% used net present value. Those with a target multiple required 5.5 times on average, and respondents reported roughly a quarter of their investments losing money.

Build the return case on your own forecast, not management's. In Kaplan and Strömberg's sample, the median management forecast of year-four earnings before interest and taxes was $11.9 million, against $6.5 million in the investors' own forecasts¹. At the example's $32 million price, the arithmetic runs like this.

1. 1

   Ownership at entry

   A $5 million check in a $10 million round at a $32 million pre-money buys 11.9% of the company.

2. 2

   Ownership at exit

   If two later rounds each sell 20% of the company, your stake falls to about 7.6%.

3. 3

   Exit value needed

   Returning 5 times your money, or $25 million, takes a sale of about $330 million, ignoring liquidation preferences.

4. 4

   The losing case

   Show what you recover if the company sells for less than the capital raised, or fails.

Read the terms from the term sheet and the charter, not from a summary: liquidation preference, participation, anti-dilution, board seats and pro rata rights. The National Venture Capital Association publishes model documents for venture financings⁵, so flag every place this deal departs from them.

A cap table and exit waterfall shows what each holder receives at different sale prices, and what each term costs. A comparable company analysis tests the price against what similar companies are worth.

End with the decisions, numbered in the order the committee must take them: the price, any exception to fund policy, and whether to vote before missing documents arrive. In the Gompers survey, roughly half of the funds required a unanimous vote of the partners⁴, so at those funds one unconvinced partner can stop a deal.

Turn every absent figure into a request that names the document, the period and the system it comes from. Send the list as soon as you find the gap, so answers arrive before the vote.

Then give the draft to a colleague who was not on the deal. If they cannot trace a number to its source or explain the price, fix the memo before the committee reads it.

The template on this page holds the memo, its three sheets, the risk register, the request list and a presentation outline. Edit with AI opens it in River, which builds the metric sheet from your deck and data room before drafting anything.