Investment Thesis in Venture Capital: Examples and Steps

An investment thesis is a written statement of what an investor will back, why those investments should outperform, and what edge the investor has in finding and winning them. In venture capital it operates at two levels: the fund thesis a general partner pitches to limited partners, and the deal thesis an investor writes to justify a specific check.

Definition: An investment thesis is a falsifiable argument that a defined set of companies, chosen by stage, sector, geography, and business characteristics, will generate returns above the market because of an insight or advantage the investor holds.

Everyone says they have a thesis. Fewer can tell you what it makes them turn down.

Our test is simple. A good thesis is specific enough to say no to most opportunities, measurable enough to check against results, and honest about why this team should be the one to execute it. Here's an illustrative example of the kind we'd want to read:

We invest $500K to $1.5M in pre-seed and seed rounds of US software companies selling to regulated industries, where compliance requirements create switching costs. Our partners spent 20 years selling into those industries and can reach buyers and founders competitors cannot. A $30M fund making 30 investments and holding about 8 percent at exit needs roughly $1.13B of combined exit value from its winners, say three exits of about $375M each, to return the fund three times.

Every clause can be checked. That's much of the point. (The math: 3 x $30M = $90M of proceeds, and $90M divided by 8 percent is $1.125B.)

Why an investment thesis earns its keep

We'd write the thesis before the first check. It does three jobs that little else in your process does.

It filters. Much of deal flow is noise. The largest survey of how venture capitalists decide, by Gompers, Gornall, Kaplan, and Strebulaev (885 VCs at 681 firms, fielded between November 2015 and March 2016 and later published in the Journal of Financial Economics), found that a firm considers roughly 100 opportunities for every deal it closes. Without a thesis, all 100 may need a look. With one, many can be declined in a sentence, and the time goes to the handful that fit.

It raises money. LPs are committing capital for ten years to a strategy they can't change, so a generic thesis can read as a manager without a plan. Carta's guide to fund theses is direct on this: a vague thesis is a major red flag in LP due diligence, and so are goals that can't be measured with concrete metrics. The current market may be raising that bar further. The NVCA's 2026 Yearbook reports only 101 first-time funds closed in 2025, the lowest annual count since 2007.

It creates deal flow. Founders in a niche look for the investor known for that niche. A clear thesis can make you that investor, which arguably matters more at seed than at later stages.

What goes into a fund investment thesis

One way to check a thesis is a four-part test we call Box, Belief, Edge, Math, plus a list of exclusions. If a part is missing, the thesis probably isn't finished.

The box. Fund size, number of companies, initial check size, reserves for follow-ons, stage, sector and geography. Name markets, not "technology." Stage often matters most because it sets diligence depth and ownership targets. VC Lab, which trains emerging managers, sets a useful discipline here: keep the thesis to 40 words or fewer (ideally 35 to 37) and name a sector most people already understand, such as fintech, digital health, SaaS, or marketplaces.
The belief. What you think is true about the market that isn't yet priced in. One sentence, stated so that it could turn out to be wrong.
The edge. Why your team sees these companies first, wins the allocation, and helps more than the next investor would. Be realistic about what LPs weigh. VC Lab ranks investment exits, investment performance, and capital raised as the top three credibility signals, ahead of sales closed, companies helped, network size, and years of experience. If you have none of the top three yet, your edge likely needs to be unusually concrete.
The math. How ownership, check size, and expected loss rate produce a fund-level return. Our venture capital portfolio strategy guide walks through the models.
The exclusions. What you won't do, written down. This is the part many drafts skip. In our view it's also the part that makes the rest believable.

Three investment thesis archetypes, and where each one cracks

Many theses fall into one of three shapes. Pick yours on purpose, because each tends to fail in a predictable way.

Sector thesis.

We back seed-stage software for the construction industry, because the sector is large and underdigitized, and our partners ran operations at two construction technology companies.

Where it cracks: the sector goes cold and the fund can't pivot.

Stage thesis.

We lead pre-seed rounds of $1M to $2M in US companies with two technical founders and a working prototype, taking 10 to 15 percent ownership.

Strength: a clear place in the market. Where it cracks: valuations rise faster than the model assumes. Carta's State of Pre-Seed: 2025 in Review reports median post-money SAFE valuation caps of roughly $10M for rounds of $250K to $1M and $15M for rounds of $1M to $2.5M. At a $15M cap, a $1.5M round buys exactly 10 percent, the bottom of the target, so a rise in caps can break the model. We'd re-check ownership targets against live data each year.

Insight thesis.

Agentic software will replace seat-based pricing with outcome-based pricing across professional services, and we back the infrastructure that makes that possible.

Strength: early to a large wave. Where it cracks: everyone else holds the same view. The NVCA 2026 Yearbook reports that AI captured 65.4 percent of US venture deal value in 2025, up from 50.9 percent in 2024, and the PitchBook-NVCA Venture Monitor puts AI at 86 percent of US venture dollars in the first half of 2026.

A view most of the market shares isn't an edge. It's the consensus. There's also a live question about which layer of AI to back, which we come back to below.

The deal thesis: making the case for one company

A deal thesis is the argument for a single investment, and it usually belongs in the first paragraph of the investment memo. We'd answer five questions, in this order:

What does this company have to become for this investment to return the fund?
Why is this team the one to build it?
Why now, and why hasn't it already been built?
What is the single biggest risk, and what would we need to see to retire it?
What price and ownership make the risk-reward work?

We lead with the team, and the survey evidence supports that weighting. In the Gompers, Gornall, Kaplan, and Strebulaev data, 95 percent of VC firms named the management team as an important factor in selecting investments and 47 percent named it the single most important one. That puts it ahead of business model (83 percent important), product or technology (74 percent), and market (68 percent).

A deal thesis with three pages on the market and one line on the founders may have the weighting upside down. Our venture capital investment memo guide shows how the thesis sits inside the full document.

How to write an investment thesis in six steps

Step 1: start with what you know better than most investors. Edge often comes from an industry you worked in, a field you trained in, or a community you belong to. Write down three things you believe about it that most people don't.

Step 2: draw the box. Stage, check size, sector, geography, and founder profile. Write the exclusions early, while you still feel like being honest about them.

Step 3: do the return math. Model the portfolio: number of companies, ownership at entry and exit, expected loss rate, and the size of the winners you need. It's prudent to assume most investments won't matter much. Andreessen Horowitz's analysis of Horsley Bridge data found that about 6 percent of investments produced roughly 60 percent of returns. Angels commonly aim for at least 20 to 30 checks so the power law has a chance to work, though the right number varies with check size, access, and how much follow-on capital you hold back. We don't think there's one right portfolio size; the thesis should set it, not habit. For how outcomes are measured, see venture capital fund performance metrics.

Step 4: name the evidence. List five to ten private companies that fit the thesis and note what they share. If you can't find them, the thesis may be too narrow or the belief may be wrong.

Step 5: write two versions. A one-paragraph version in under 60 words, then a two-page version with the evidence and the math.

Step 6: stress-test it, then publish it. Which of your last ten passes would this thesis have told you to take? Would a competitor using the same words have the same edge? If both answers hold up, put it in public (more on why below).

"But publishing a thesis just hands competitors your playbook"

Some investors keep their thesis private for exactly this reason. Being wrong in public is embarrassing, and a sharp idea in a blog post can be copied by a bigger fund with a bigger check.

But.

A thesis in a drawer filters your own pipeline. A published one also shapes who enters it. Many of the founders you most want to meet haven't launched or started fundraising yet, and a public post tells them which investor to call first. Timing is much of the value.

Michael Mignano, a general partner at USV who co-founded Anchor, which Spotify acquired, made this case in July 2026. USV's June 2026 post on what it labels the Rebel Alliance (a full stack of models, including open-weights models, plus orchestration, harnesses, memory, identity and payments, built by many companies competing and composing with each other) is a working example of a firm taking a position in the open.

As for being wrong in public: investing is placing bets. A thesis bet can be hedged with a counter-position elsewhere in the portfolio. We think the cost is usually worth paying.

Where we land: publish the markets, not the roadmap

Our view is that a seed thesis is usually worth making public, and that it works best describing markets rather than products.

Keep it about the market, not the build. Former founders can be especially prone to this one. They decide how a product should work and push that view onto the companies they back. Your thesis can say which problems, markets, and business characteristics you back and why you have an edge. In our view it shouldn't dictate the roadmap. A thesis that only admits companies building your preferred product is a hidden operating plan, and it screens out founders who see the market more clearly than you do.

Be precise about which layer of AI you back. The capital-heavy build-out of AI infrastructure has produced the technology, and a large wave of applications is now being built on top, much as the internet's applications followed fiber and broadband. That argues, in our view, for a sharp thesis: there may be so many applications that it's hard to pick well without knowing exactly what you're looking for. Serious investors disagree on timing, though, and value will still accrue at the infrastructure layer, so an infrastructure thesis remains viable. Pick the layer where your edge applies and say why. Our guide to AI startup moats covers what now counts as defensible at the application layer.

For an application-layer belief stated precisely enough to act on, consider Mignano's backing of Suno, the AI music company, while at Lightspeed. In his view Suno unlocks a new consumer behavior, creating music purely for entertainment, much as YouTube and TikTok opened up video creation.

That's our answer. A fund with a strategic LP base or a sensitive sector may reasonably choose differently.

Common investment thesis mistakes

Too broad. "Early-stage technology companies with great teams" describes nearly every fund and therefore no fund in particular.
No edge. A sector focus without a reason you win the deal is a preference, not a thesis.
Math that doesn't close. A $50M fund that holds 5 percent at exit needs $1B of exit value to return the fund once and $3B to return it three times. If the sector hasn't produced an outcome that size, the thesis likely needs rework.
A roadmap in disguise. If only one product design qualifies, you've written an operating plan for someone else's company.
Never revisiting it. Check the thesis against the actual portfolio each year during the venture capital portfolio review.

Writing your thesis as a new angel

In our view, angels need a thesis as much as funds do. Without one, it's easy to write checks based on who asked rather than what you believe. One approach is a paragraph covering your box and your edge, a set check size and number of investments (a $10K check across 25 companies is a $250K plan), and a decision in advance about what you won't fund.

A thesis becomes real the first time you use it to say no. For accredited investors new to angel investing, 1752vc's Emerging Angels program is an 8-week live program that gives them a seat in a working fund's investment process: live diligence calls and deal reviews, plus monthly Investment Circles and a private community. That's where a written box gets tested against companies other people are also arguing about. Anyone aiming at the fund side can instead source against a thesis in the Venture Fellow program, where about half of 1752vc's deal flow is sourced by Fellows.

For founders: use the framework in reverse. Read a fund's thesis before pitching and lead with the part of your company that fits it. See the venture capital hub for more.

The bottom line

A thesis is less about what you believe than about what it lets you ignore. If it has never made you pass on a company you liked, it hasn't started working yet.

Conviction gets you into a deal.

A thesis gets you out of the other ninety-nine.

Key takeaways

An investment thesis is a falsifiable argument about which companies an investor backs, why they will outperform, and what edge the investor has.
It filters the roughly 100 opportunities a firm considers per closed deal, persuades LPs, and attracts the right founders.
A complete fund thesis can be checked with a simple Box, Belief, Edge, Math test and a list of explicit exclusions.
Deal-level theses live at the top of the investment memo and in our view lead with the team, which surveyed VCs most often rank as the most important factor.
It helps to test a thesis against the power law, since a few investments may need to return the entire fund.
We suggest publishing a seed thesis so founders find you before they launch, and keeping it about markets rather than how founders should build.

Frequently asked questions

What is an investment thesis in venture capital?

It is a written argument that defines which companies a fund or investor will back (by stage, sector, geography, and business characteristics), why those companies should generate above-market returns, and what advantage the investor has in finding, winning, and helping them. LPs use it to decide whether to commit, and investors use it to decide what to pursue.

How do you write an investment thesis?

A common approach: start with what you know better than other investors, draw the box (stage, check size, sector, geography, exclusions), model the portfolio math so a few winners can return the fund, list real companies that fit, and compress it into a paragraph. Then stress-test it against your past decisions and revise it each fund cycle.

Do angel investors need an investment thesis?

In our view, yes. A thesis tells you what to say no to, sets a check size and portfolio count that give the power law a chance to work, and keeps you from investing based on who asked. One paragraph with a defined box, an edge, and exclusions is enough to start.

Why do LPs care about an investment thesis?

Because they are committing capital for ten years to a strategy they cannot change. Carta's guidance on fund theses treats a vague or generic thesis as a major red flag in due diligence, along with goals too vague to measure. With only 101 first-time funds closed in 2025 per the NVCA 2026 Yearbook, the lowest count since 2007, the thesis can be a deciding factor for a new manager.

Should VCs publish their investment thesis?

We think seed investors usually should, though some prefer privacy. A public thesis tells founders who have not yet launched which investor fits them, and being wrong in public is a cost you can hedge across a portfolio. We would share the markets you back and your reasoning, but not a prescription for how founders should build, because that can screen out founders who see the market better than you.

Sources

Disclaimer: This guide is for general education only and is not legal, tax or investment advice. Laws, market data and program terms change, so it may not reflect the latest developments or fit your situation. Treat it as a starting point, not a source of truth, and talk to a qualified lawyer, accountant or financial adviser before you make decisions.