How to build deal flow that actually converts

A quiet inbox isn't a sign of good taste. It's a sign that founders don't know you exist yet.

Deal flow is the only part of investing you can directly control. You can't control whether a company will be a 10x. You can't control market cycles. But you can control how many good founders find their way to you, and how seriously they take you when they do.

Here's what actually works, and why the old arguments against running high volume no longer hold.

Where deals come from

The research here is unambiguous. A survey of 885 institutional VCs by Gompers, Gornall, Kaplan, and Strebulaev, published in the Journal of Financial Economics and summarised in Harvard Business Review (2021), found that roughly 60% of deals trace back to professional networks, co-investor referrals, or portfolio company introductions. Not cold outreach. Not pitch competitions. Relationships.

That number matters for two reasons. First, it tells you where to spend your time. Second, it tells you what kind of reputation you need: one that makes other investors and founders want to send you their best companies, not bury you in stuff they're not excited about.

The same study found that for every investment a VC firm makes, it evaluates an average of 101 companies. Roughly 28 of those get a meeting. Ten go through serious due diligence. Four receive a term sheet. One closes.

That 1-in-101 ratio isn't a quirk of large US mega-funds. It's the math of early-stage investing. You need volume at the top of the funnel to find quality at the bottom.

The five channels that work

Your existing network. Co-investors, angels, lawyers, other GPs: these are the people who see deal flow you don't. One genuine relationship with an active angel in your space is worth more than six months of cold LinkedIn outreach. The goal is to be the name they think of when a founder asks "Who else should I be talking to?"

Your portfolio companies. Founders know other founders. If your portfolio companies believe in what you're doing, they'll refer their network. This is the highest-conversion source for most funds, because the founders coming in already have social proof attached.

Accelerators and incubators. Programs like Sting, SSE Business Lab, Ignite, or regional incubators graduate cohorts of screened, stage-appropriate companies. Building even one or two relationships with program managers there can generate a reliable flow of warm introductions. They want to know which investors take their companies seriously.

Content and thought leadership. Writing about what you believe (your thesis, your take on specific sectors, the questions you keep asking in founder meetings) attracts founders who are already aligned. A blog post about why you invest in a specific kind of business works as a permanent pitch to every founder who reads it. It also sets expectations upfront, which means fewer misaligned applications wasting everyone's time.

A public application link. Make it easy to apply. Not buried in your website footer: share your fund's application URL in your bio, your email signature, and any public context where founders might find you. Some of the best companies in any fund came through a direct cold application from a founder who just figured out they wanted that specific investor.

None of these channels work in isolation and none of them work overnight. The funds with the best deal flow treat it as ongoing work, not something you set up once.

Volume is not vanity

The 1-in-101 funnel isn't a problem to solve. It's the nature of the asset class.

If you want to close three investments this year, you need around 300 companies to enter the top of your funnel. If you're evaluating 30, you're not being selective. You're just not seeing enough.

The old counterargument was: more volume means more work. That was true when screening meant analyst hours: reading every deck, doing background research, writing up assessments. Most lean teams could realistically handle 100 to 200 applications a year before it became noise.

That constraint doesn't exist the same way anymore.

What changes with AI screening

Auryn runs 10 to 15 parallel research tasks on each application: thesis alignment, founder background verification, market sizing, comparable companies, early red flags. The output is a structured assessment ready when you open it. The average processing time is 10 to 15 minutes per application.

At that rate, volume is no longer the bottleneck. A two-person team can run a pipeline that would have required a full analyst bench two years ago.

This changes what deal flow building is worth. If every application you receive gets a serious, consistent evaluation (not a 30-second skim before you move on), you can open the top of your funnel wide without the fear that you'll drown in it.

The public application link stops being a liability and becomes an asset. Running an active content strategy stops feeling like marketing overhead and starts looking like a high-return sourcing channel. Sharing your thesis publicly attracts founders who fit it. With AI handling initial screening, the cost of a misaligned application is near zero.

Where to start

If your deal flow is slow right now, the highest-return moves are:

Make sure your application link is live and easy to find. Auryn generates a public application page for your fund — that URL is your top-of-funnel. Put it in your bio, your email signature, and anywhere your name appears publicly.

Write something. One post about your thesis, your sector view, or a question you keep asking in founder meetings. It doesn't have to be long. It just has to be specific.

Message three people in your network this week: co-investors, accelerator contacts, or portfolio founders. Ask what they're seeing. Not to pitch them, just to stay in the conversation. The deals worth doing often come with two degrees of separation, not zero.

None of this requires a big operation. It requires consistency at a small scale.

The top of the funnel is where you have the most control. Use it.