An investment memo is the document a fund decides from. It is also the document a fund learns from, years later, when someone asks why it backed a company or passed on it. A memo that reads like a pitch deck in paragraph form does neither job well. It repeats the founder's story without showing what the fund checked, and it hides the risks the fund chose to accept.
Learning how to write an investment memo that works means writing for a sceptical reader who has twenty minutes. The memo should make the recommendation easy to find, the evidence easy to check and the risks impossible to miss. The sections below cover structure, evidence, risks, style and the mistakes that weaken most memos.
What a memo is for, and what research says it covers
A memo has three readers. The committee, who decide from it. The sponsoring partner, who is forced by writing it to separate conviction from evidence. And the future reader, who wants to know what the fund knew at the time.
Research on venture memos gives a sense of what they usually contain. Kaplan and Strömberg studied the internal investment analyses that 11 venture firms prepared for 67 portfolio investments. The analyses covered the attractiveness and risks of the business, the management team, the deal terms and the monitoring the firm expected to do after investing. As summarised by Gompers, Gornall, Kaplan and Strebulaev, the memos focused on the team, the market or industry, the competition, the product or technology and the business model, without ranking those criteria against each other.
That last point is worth acting on. A good memo does rank. It says which two or three factors drive the recommendation, so the committee can argue about the right things.
How to write an investment memo, section by section
Use the same structure for every memo so partners know where to look. A structure that works for most early-stage deals:
| Section | What it answers | Typical length |
|---|---|---|
| Recommendation | Invest, pass or invest with conditions, the amount, and the two or three reasons | A short paragraph |
| Company in one paragraph | What the company does, for whom, and why now | One paragraph |
| Team | Why these founders for this problem, and the gaps | Two or three paragraphs |
| Market | Who buys, how many of them, and how the company reaches them | Two or three paragraphs |
| Product and traction | What works today, with each metric defined and sourced | Half a page |
| Thesis fit | Which criteria are met, which are unknown, and any exceptions | A short list |
| Terms | Round size, instrument, valuation or cap, ownership, special rights | A short list |
| Risks and what must be true | The main risks, the evidence on each, and what you are accepting | Half a page |
| Open questions | What diligence could not answer and how you would monitor it | A short list |
Sequoia's guide to writing a business plan lists purpose, problem, solution, why now, market, competition, business model, team, financials and vision. Those are the founder's headings. The memo's headings are the fund's: they add evidence, fit, terms and risks, and they lead with the decision. Our investment memo template has this structure ready to fill in.
Separate checked facts from founder claims
The most useful habit in memo writing is to say, for every material number, where it came from and whether anyone checked it. A memo that states "revenue is growing 20% a month" without a source tells the committee nothing about how far to trust it.
- Cite the source of each material fact: the deck slide, the data room file, the customer call, the interview answer.
- Label its status: reported by the founder, corroborated by a document or call, disputed by another source, or still unresolved.
- Write the definition next to the number. "ARR" built from one month multiplied by 12 is not the same as contracted recurring revenue.
- Keep earlier statements when facts changed. If the deck said 20 customers and diligence found 12 paying customers and 8 unpaid pilots, show both and explain the difference.
- Do not round away uncertainty. "About 40 customers, of which 15 confirmed" is more useful than "40 customers".
Our guide on how to verify startup traction covers how to check the numbers before they go into the memo.
Write the risks honestly
Every investment carries risks. The memo's job is to name the ones that matter, show the evidence on each, and say plainly which ones the fund is choosing to accept. A useful framing is "what must be true" for the investment to work.
- 1List three to five things that must be true, such as "mid-sized clinics will pay for this without a reimbursement code".
- 2For each, give the evidence for and against, with sources.
- 3Say whether diligence could test it, or whether only time will tell.
- 4Name the signal you will watch after investing, and when you expect to see it.
- 5Include the strongest argument for passing, written as fairly as the case for investing.
Avoid a generic risk list that could apply to any startup, such as competition, execution and fundraising. Specific risks lead to specific questions in committee and specific conditions on the investment.
Style and common memo mistakes
- Burying the recommendation. Put it in the first paragraph, not the last page.
- Writing a pitch. The founder already wrote one. The memo is the fund's assessment, so it should include doubts.
- Length as a proxy for rigour. Most early-stage memos work at four to six pages plus appendices. Move raw data and call notes to the appendix.
- Adjectives instead of evidence. Replace "exceptional team" with what the founders have done that is relevant to this company.
- Judging presentation rather than substance. Note what founders said and showed, not how polished they sounded.
- No dates. Date the memo and each data point, because a memo is a snapshot of what the fund knew at that time.
Once the memo is done, it goes into the committee packet. Our guide on how to run an investment committee covers what happens next.
Where Delsivo fits
Memo drafting is part of the committee workflow planned for a later Delsivo release. The design is for AI to draft a memo from the deal record, with every statement linked to its source, for the sponsoring partner to edit and approve. Quotes that cannot be found in the source are dropped rather than shown.
The record a memo would draw on is part of the pilot design: deck facts with slide citations, a thesis screening with each criterion's source and result, and a claim tracker where each claim keeps its definition, period, units and evidence label. AI suggests and drafts. The partner writes the recommendation and the committee decides. See the claim tracker for how claims are recorded.
Sources
- Kaplan and Strömberg: Characteristics, Contracts, and Actions: Evidence from Venture Capitalist Analyses (NBER Working Paper 8764, February 2002) (checked 6 November 2026)
- Gompers, Gornall, Kaplan and Strebulaev: How Do Venture Capitalists Make Decisions? (NBER Working Paper 22587, September 2016) (checked 6 November 2026)
- Sequoia Capital: Writing a business plan (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.