Every fund sources deals through some mix of channels, whether it plans that mix or not. A partner's network sends a few warm introductions a week. An application form on the website collects a steady stream of cold submissions. An associate builds lists of companies in a sector and writes to founders. A scout passes on a company from their university.
Thinking clearly about inbound vs outbound deal flow helps a lean team decide where to spend its limited hours. Each channel produces different companies, at a different cost in time, with different risks of missing good founders. The sections below look at what survey data says about where deals come from, compare the channels, and set out how to run and measure each.
Where venture deals actually come from
The best public data on sourcing comes from the survey of 885 venture capitalists at 681 firms by Gompers, Gornall, Kaplan and Strebulaev. They asked where the deals each firm closed in the previous twelve months had originated. Over 30% came through professional networks, about 20% were referred by other investors, 8% were referred by portfolio companies, almost 30% were proactively self-generated by the firm, and about 10% came inbound from company management. Quantitative sourcing, using data to identify companies, accounted for about 2%.
Stage made a difference. Early-stage firms closed more inbound deals (12%) than late-stage firms (7%), and more deals referred by portfolio companies. Late-stage firms generated far more of their deals proactively (42%, against 23% for early-stage firms). The authors also found little difference in pipeline sources between more and less successful firms, and suggested that the quality of a firm's referral network may matter more than the type of source.
Two cautions. These figures describe closed deals, not the whole pipeline, so inbound may be a much larger share of what a fund sees than of what it backs. And the survey dates from 2016. Treat it as a picture of the shape, not a benchmark for your fund.
Inbound vs outbound deal flow compared
Each channel has a different profile. Most funds need all three, in proportions that fit their stage and team.
| Aspect | Inbound | Outbound | Referred |
|---|---|---|---|
| How it arrives | Applications, cold emails, events, content | Your own research lists and outreach | Introductions from your network, other investors, portfolio founders, scouts |
| Volume | Can be high and uneven | As high as your team's capacity | Moderate, depends on network size |
| Time cost per deal | Low to read, high in total at volume | High: research and outreach before any conversation | Low: someone has filtered already |
| Fit with thesis | Mixed unless your thesis is published | High, because you chose the targets | Depends on how well referrers know your thesis |
| Who it reaches | Founders outside your network who know you exist | Founders you identify, often less visible ones | Founders already close to your network |
| Main risk | Slow or missing replies damage your reputation | Narrow research recreates the same picks as everyone else | Pipeline mirrors your network and misses others |
The last row matters most. Every channel has a blind spot. Referrals are efficient but repeat the shape of your network. Outbound finds what you look for and nothing else. Inbound depends on founders knowing you exist and believing you will read their application.
Run inbound well
Inbound gets a bad reputation because many funds run it badly: a contact form that goes to a shared inbox nobody owns. Run properly, it is the channel most open to founders without connections. Accelerators show how far an open process can go. Y Combinator's FAQ describes an application process open to any startup anywhere in the world, and notes that in a typical batch about half the companies applied more than once before being accepted.
- 1Publish what you back and what you do not, so founders outside your thesis can skip applying. Our guide on how to write an investment thesis covers the founder-facing part.
- 2Use a proper application form that asks only for what you will use at the first screen, with the deck as an upload.
- 3Assign an owner for the inbound queue each week.
- 4Screen against written criteria, and treat missing information as a question rather than a failure.
- 5Commit to a reply time and keep it, including for declines.
Our guide on how to screen startup applications covers the screening process in detail.
Run outbound and referrals deliberately
Outbound
- Start from your thesis: pick a sector or problem where you have a view, and list the companies working on it.
- Write to founders with something specific: why their company caught your attention and what your fund brings.
- Record every outreach as a deal with its source, even before a reply, so you can see what the effort produces.
- Expect most outreach to be early conversations, not live rounds, and track when to follow up.
Referrals and scouts
- Tell referrers what you are looking for, briefly and often, so the deals they send fit.
- Thank referrers and tell them what happened, including passes. People keep referring to funds that close the loop.
- Give scouts and frequent referrers a way to send deals that keeps the source attached, rather than forwarded emails.
- Watch for a narrow network. If most referred deals come from a handful of people, your pipeline looks like them.
Measure each channel by source
You cannot manage a mix you do not measure. Record the source of every deal when it first arrives, and keep it when the same company arrives again through another channel. Then, once a quarter, look at each channel's numbers.
- Volume: how many deals each source produced.
- Fit rate: what share passed your hard rules.
- Progression: what share reached a first meeting, diligence and committee.
- Time cost: roughly how many team hours each channel used.
- Reply time: whether inbound founders heard back within your committed time.
Do not judge a channel on one quarter, and do not cut inbound just because its fit rate is low. A low fit rate often means your thesis is not published clearly, which is fixable. Common mistakes: no source field at all, overwriting the source when a company arrives twice, and comparing channels on volume alone. Our guide on VC deal flow management covers the wider pipeline.
Where Delsivo fits
Delsivo is designed to make inbound and referred deal flow easier to run in one record. In the pilot release, founders apply through your branded portal on your own subdomain, through private referral links that keep the source attached, or an analyst creates the record for outbound and referred companies. Possible duplicates are suggested by company domain and founder identity, and an analyst confirms any merge so the original source is not lost. Each deck is screened against your thesis, with Unknown turned into a question rather than a failure.
An intake mailbox for forwarded pitches, scout links, CRM import and export, and relationship mapping are planned for a later release. People decide every stage change and pass, and approve every message to a founder. See the startup application portal and VC scout program software pages.
Sources
- Gompers, Gornall, Kaplan and Strebulaev: How Do Venture Capitalists Make Decisions? (NBER Working Paper 22587, September 2016) (checked 6 November 2026)
- Y Combinator: Frequently asked questions (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.