How Venture Capitalists Make Investment Decisions: The Data

Venture capitalists typically make investment decisions by running each opportunity through a steep funnel, weighting the founding team above the product or market, and testing whether the company could plausibly return their entire fund. In the most detailed survey of the process, a firm considers roughly 100 opportunities for every deal it closes, and the management team ranks as the single most important factor in both the decision to invest and the eventual outcome.

That survey, of 885 venture capitalists at 681 firms, was run by Paul Gompers, Will Gornall, Steven Kaplan, and Ilya Strebulaev between November 2015 and March 2016. It was published as NBER Working Paper 22587 and then in the Journal of Financial Economics in 2020. It also found that the average deal takes 83 days from first meeting to close.

Our read: the process is more structured than founders assume, and more judgment-driven than the spreadsheets suggest.

How venture capitalists make investment decisions: the funnel

Most opportunities are cut early. In the Gompers, Gornall, Kaplan, and Strebulaev survey, for every 100 opportunities a venture firm considers:

About 25 lead to a meeting with management.
About 8 are reviewed by the partnership (roughly a third of the meetings).
About 4 go into due diligence (roughly half of the partner reviews).
1 to 2 receive a term sheet: the survey reports about 1.7 term sheet negotiations for every deal a firm closes.
Roughly 1 closes, a close rate near 60 percent, because competing firms bid for the same deals at the same time.

The median firm closes about four deals a year. IT-focused firms considered around 151 deals per investment, while healthcare firms considered around 78, reflecting the higher cost of evaluating a scientific company.

Where the 100 come from matters as much as how they're cut. The same survey found about 30 percent of deals came from the VCs' professional networks, 30 percent were self-generated by the investors, 20 percent were referred by other investors, about 10 percent were inbound from company management, and 8 percent came from portfolio companies. Warm paths dominate in that data. That's why our venture capital deal sourcing guide spends so much time on network building.

What VCs weigh: team first, then business model, product, and market

Ask a venture capitalist what they look for and a common answer is "the people." The survey broadly backs that up. When VCs were asked which factors were important in selecting investments, 95 percent named the management team, 83 percent the business model, 74 percent the product or technology, 68 percent the market, and 31 percent the industry. Forced to pick the single most important factor, 47 percent chose the team, while all business-related factors combined were ranked first by about 37 percent.

The pattern holds after the fact. In the same survey, 96 percent of VC firms cited the team as a contributor to their successes and 92 percent cited it as a contributor to their failures. Timing, luck, technology, business model, and industry conditions were each named far less often.

For founders: the first meeting is a test of you before it's a test of the deck. Within "team," respondents put managerial ability, relevant industry experience, and passion at the top, so show all three. We've written about how an investor runs that first meeting, and it's worth seeing from the other chair.

For new angels: reference calls are often more valuable diligence than spreadsheet work.

The metrics VCs actually use

The financial toolkit in venture is simpler than in private equity, and the survey suggests it's used less rigorously than you might expect.

Cash-on-cash multiple is the most common metric, used by about 63 percent of VCs in the survey. The median required multiple is 5x (the average is about 5.5x).
IRR is used by about 42 percent, with an average required IRR of 31 percent. The paper notes that late-stage and larger VCs require lower IRRs of 28 to 29 percent, while smaller and early-stage VCs require more.
NPV is used by only about 22 percent.
About 9 percent of VCs said they use no financial metrics at all, rising to 17 percent among early-stage investors. Roughly 20 percent of firms don't forecast company cash flows, and 31 percent of early-stage VCs don't; among those who do, the median forecast period is three to four years.

In our view, the calculation that matters most isn't on that list. It's the fund-return test: a $100M fund that will own 10 percent after dilution needs a $1B exit to return the fund once.

Chris Dixon's 2015 analysis of Horsley Bridge data for Andreessen Horowitz helps explain why many VCs lean on that math. About 6 percent of investments, representing about 4.5 percent of dollars invested, produced roughly 60 percent of total returns across the funds studied since 1985. Nobody knows in advance which company will be the outlier, so many investors want each one to have a credible path to being it. That's the power law at work, and it explains a lot of passes that feel personal but aren't.

The venture capital method valuation guide shows how VCs back into a price from a target multiple.

How venture capitalists make investment decisions step by step

Most institutional firms follow a version of this sequence, whatever they call each stage.

Screen. An analyst or associate checks the deck against the firm's thesis and stage and decides whether to take a meeting. Many deals end here.
First meeting. Usually 30 to 60 minutes. The goal is to decide whether the team is worth the firm's time, not to price the deal.
Partner review. The lead sponsor presents the opportunity to the partnership, often at a Monday meeting, and asks for permission to spend diligence resources.
Due diligence. Customer calls, founder references, market analysis, product review, financial and legal checks. The survey found the average firm spends 118 hours on due diligence and calls 10 references per deal. Founders can prepare for those calls with our guide to how VCs run founder reference checks. See our venture capital due diligence guide for what gets checked.
Investment memo. The sponsor writes a document that argues for the investment, states the risks plainly, and models the return case. The venture capital investment memo guide shows the structure.
Investment committee. The partners decide, and it's worth asking early how. There is no legal standard: firms set their own rule, ranging from a broad veto (Sequoia partner Shaun Maguire has described one where everyone holds a veto, though vetoes are rare) to a champion rule (at Khosla Ventures, per Vinod Khosla, one believing partner can proceed over objections), with simple or supermajority votes and consensus discussion in between. See venture capital investment committee.
Term sheet and close. Negotiation, confirmatory diligence, legal documents, and wire. The survey puts the average deal at 83 days from first meeting to close.

The terms VCs say they rarely bend on

Decision-making doesn't stop at yes. The survey asked investors which deal terms they consider flexible. VCs described themselves as relatively inflexible on pro rata rights, liquidation preferences, anti-dilution protection, vesting, valuation, and board control. They were more flexible on the option pool, participation rights, investment amount, redemption rights, and especially dividends.

Use that list as a map for negotiation. Pushing hard on liquidation preferences or pro rata rights may spend goodwill for little return. Asking for a change to the option pool is asking for something the investor can more easily give. Our guide to liquidation preference explains why the most protective term sits firmly on the inflexible side.

Carta's State of Private Markets: 2025 in Review found that fewer than 14 percent of new fundings in Q4 2025 were down rounds, the lowest rate in three years, and that median dilution on all rounds from seed through Series C had fallen from about 18 percent to 16 percent. One reading is that investors are paying more for the same ownership, which may sharpen their focus on the terms that protect them if valuations fall.

"But that survey is ten years old"

Fair. It was fielded in 2015 and 2016, before today's AI tools and at least one full boom and bust. Some numbers, like the 83 days or the 118 hours of diligence, have probably moved.

But the shape is harder to change than the numbers. A fund can still only back a handful of companies a year, still needs a few outliers to carry it, and still bets mostly on people. We haven't found a newer study that looks at the process in this depth, so we'd treat the figures as a baseline and the logic as the durable part.

Why smart investors still get it wrong

In our view, the process is better at avoiding bad companies than at recognizing unfamiliar great ones. CB Insights' collection of investor accounts of missed deals makes the point well. Fred Wilson of Union Square Ventures has written that the firm couldn't wrap its head around air mattresses on living room floors as the next hotel room. John Greathouse of Rincon Venture Partners worried Uber had to build both sides of a marketplace at once. Kevin Rose passed on Pinterest at a $5M valuation before he joined Google Ventures and later called it the one that got away. Each pass was defensible on the day.

Three biases seem to recur: pattern-matching to past winners, anchoring on last year's prices, and judging the market as it is rather than as it could be. Many good firms try to counter them with memos that state what would need to be true, post-mortems on passes as well as losses, and "anti-portfolio" tracking of the deals they declined.

Learning the process from the inside

A funnel described is not a funnel run. 1752vc's Venture Fellow program moves people through the second version over eight weeks of live virtual sessions, working case studies and real pitch materials and then diligence on live companies, so the screen to meeting to memo sequence happens on companies that are genuinely raising. It is built for aspiring VCs, professionals moving into investing, and founders who want to understand how investors decide, and applications are reviewed on a rolling basis.

The bottom line

VC decisions look like a process from outside and feel like a judgment call from inside. Both are true. The funnel decides what gets a look; the team and the fund math decide what gets a check.

A pass is usually about the portfolio.

A yes is usually about the people.

Key takeaways

Venture capitalists make investment decisions through a steep funnel: for every 100 opportunities considered, about 25 management meetings, 8 partner reviews, 4 diligence processes, and roughly 1 closed deal, according to the Gompers, Gornall, Kaplan, and Strebulaev survey of 885 VCs at 681 firms.
In the survey, the team is the dominant factor: 95 percent of VC firms call it important and 47 percent call it the single most important, and 96 percent of firms name it as a contributor to their successes and 92 percent to their failures.
Cash-on-cash multiple (median required 5x) and IRR (average required 31 percent) are the main metrics, but in our view the fund-return test often drives the real decision.
VCs describe themselves as inflexible on liquidation preference, pro rata, anti-dilution, vesting, valuation, and board control, and flexible on option pool, participation, dividends, and check size.
Even disciplined processes miss great companies; written memos, pass post-mortems, and anti-portfolio tracking are common ways good firms try to improve.

Frequently asked questions

How do venture capitalists make investment decisions, and what matters most?

They run each opportunity through a funnel of screening, meetings, partner review, diligence and a committee decision, and they weight the founding team above everything else. In the Gompers, Gornall, Kaplan, and Strebulaev survey of 885 VCs, 95 percent named the management team an important factor and 47 percent named it the most important, ahead of business model, product, and market.

How long does it take venture capitalists to make investment decisions?

The average deal took about 83 days from first meeting to close in the Gompers, Gornall, Kaplan, and Strebulaev survey, with about 118 hours of due diligence and 10 reference calls inside that window. That is an average, not a rule: competitive rounds close much faster and complicated ones take far longer.

Why do VCs pass on most startups they meet?

Usually because a fund only has room for a handful of positions a year and each one ideally has a credible path to returning the fund. The median firm in the Gompers, Gornall, Kaplan, and Strebulaev survey closed about four deals a year, so a pass is usually a statement about portfolio math and fit with the thesis, not a verdict on the business.

What metrics do venture capitalists use to evaluate startups?

In the Gompers, Gornall, Kaplan, and Strebulaev survey, 63 percent used a cash-on-cash multiple, with a median required multiple of 5x, and 42 percent used IRR, with an average required IRR of 31 percent. Only 22 percent used NPV and about 9 percent used no financial metric at all. In our view, the decisive calculation is often whether the company could return the whole fund.

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.