How to run a seed fundraising process: list, sequence, close

Run a seed raise as one short, parallel process with a start date and a stop rule. DocSend’s 2024 data puts the average pre-seed raise at 12 weeks, and most successful seed companies closed in 12 weeks or less. Size the round to your next fundable milestone, usually 12 to 18 months out (Geoff Ralston, YC). Filter funds on stage, check size, sector, and whether they are writing new checks. Then meet them in parallel, ranked by how likely each one is to say yes and how much that yes is worth (Paul Graham). Close the first commitment fast and use it to pull in the rest. Not legal, tax, or investment advice.

Filter the list before you sequence it

Filters from DocSend (Heddleston, 2021: actively investing, interested in your sector, funding companies your size), Graham’s How to Raise Money (leads, speed, introductions), and YC’s seed guide (know your audience). The drop rules are editorial.

Fund size sets check size. In Carta’s Q1 2026 fund report (2,775 US funds), about 89% were smaller than $100 million, and most funds closed were smaller than $25 million. A small fund writes small checks, so match the round to the fund before the meeting. Deployment pace matters too. Funds from Carta’s 2018 to 2020 vintages deployed at least half of their capital in their first two years. The 2022 vintage slowed to 43% after two years (Carta, September 2024). Ask where the fund sits in its deployment period. The fund’s lens, once you are in the room, lives on how seed VCs decide.

Size the round to the milestone

Ralston’s rule: raise enough to reach the next fundable milestone, usually 12 to 18 months out, and tie the ask to a believable plan. Most seed rounds dilute up to 20%; try to stay under 25%. Build more than one plan. Graham describes the shape: with a little money you hire one or two people, with more you hire a team, and each plan matches a different investor.

Start low and raise the target as money commits. Graham’s example: if you want $500k, say $250k at first. At $150k you are more than half done, which signals momentum and tells investors the room is filling. Opening at $500k leaves you under a third done at the same point.

Check the round label with the investors you plan to pitch. Carta’s Q2 2026 pre-seed data shows few pre-seed deals above $2.5 million, and the ones that exceed it typically stack ten or more instruments. A FounderNexus session flagged the label itself: a “seed” label can set expectations for the next round that your plan does not support yet. Ask investors what evidence they expect behind this label and the next one, then use one label across the deck, outreach, and records. That is session judgment, not a survey. Model the months in the runway calculator and pick the paper on SAFE vs priced round.

Sequence the outreach

Graham’s core rule is breadth-first search weighted by expected value. Talk to investors in parallel. Serial meetings take too long and remove the pressure other investors create. Expected value is the chance an investor says yes times how good that yes would be. An eminent investor who is hard to convince can rank level with an easy angel who writes less. An angel who writes small checks and wants many meetings ranks last.

A note in Graham’s essay, credited to Jeff Byun, covers the warm-up. Some founders schedule a handful of less likely investors first to work the bugs out of the pitch. Packing meetings too close together also leaves less room for the pitch to improve between them. The session operators gave the same order: start with a few relevant conversations, then approach the investors whose involvement would matter most. Treat the practice meetings as real chances to raise.

Intros decide who takes the meeting. YC calls a warm introduction by far the best way to meet an investor. Graham ranks them: best from a well-known investor who has invested in you, next from a founder the investor has funded. Once someone commits, ask them to introduce you to investors they respect. Graham passes on Joshua Reeves’s version: ask for two. Skip intro requests from investors who passed. Graham warns those can read as an anti-recommendation.

Know where you stand

Graham: never leave a meeting without asking what happens next. What else do they need to decide, who else must meet you, and by when? Ralston adds the same close: leave with an attempted close or clear next steps. Treat each investor as a no until you hold a definite offer with no contingencies.

Funds also need time of their own. In the Gompers et al. survey, early-stage firms reported 73 days to close against 106 for late-stage firms (Table 6, survey run 2015 to 2016). Build that into your runway.

Close the money and protect the company

Graham: getting the first substantial offer can be half the total difficulty of fundraising. A $50k check from a well-known VC firm or angel usually gets things moving. After the first money is in, each close gets faster and easier (Ralston). Close committed money the same week. Graham hears founders say they have raised $800,000 when none of it is in the bank. On a SAFE, Ralston notes, signing and wiring can take minutes.

Accept good offers when they arrive. Graham calls this accepting offers greedily: take an acceptable offer instead of waiting for a better one that may not come. If one offer would fill most of the round, tell the others and give them a few days.

Protect the business while you raise. Graham recommends that one founder, the CEO, runs the raise so the others keep building. He also notes investors read the trend between meetings: growth between two meetings makes them keen to close, and flat numbers cool them off. The session operators warned that sales stall during a long raise. Name owners for customer follow-up and delivery before outreach starts.

Set the stop rule before you start. Heddleston’s rule of thumb: if 30 real meetings all say no, change the deck or make more progress before you contact another 30. Graham’s version: stop when you start getting air in the straw.

After a pass, send material-change notes

Graham: investors who reject you are some of your warmest leads for future rounds, and any investor who spent real time deciding probably came close to yes. Ralston: part on the best possible terms, since a no today can be a yes later. The session operators repeated one tactic more than any other: a short note to passed investors on each material change, with no ask.

Template is editorial. Keep investor updates on the same metric definitions you use in the deck and the model.

Run the seed process

Mistakes that stretch a seed raise

Serial meetings. One investor at a time takes longer and removes the pressure a parallel process creates (Graham).

Opening with followers. Investors who wait for a lead cannot give you the first yes. Graham puts their early expected value at zero.

Half in, half out. Graham: fundraising becomes the top idea in your mind, and growth drops when founders look away. Pick a window and run it.

Counting soft yeses. Verbal interest is not money. Get commitments confirmed and wired (Graham).

More contacts instead of a better pitch. DocSend finds more contacts bring more meetings, but the link to dollars raised is weaker. After a run of nos, change the deck or the progress (Heddleston).

A label your plan cannot support. The label sets the bar for the next round. Check it with investors first (FounderNexus session).

Pair with fundraising siblings

Learn what each fund weighs on how seed VCs decide. Build the deck on seed pitch deck. Close early checks on first angel check. Pick the paper on SAFE vs priced round. Read the lead’s terms on term sheet red flags. Orient the rest from the fundraising hub.

Sources

• DocSend, VCs prioritize people in an AI-heavy landscape, according to new Dropbox DocSend report (PR Newswire) — 18 December 2024. Pre-seed and seed 2024 reports, more than 400 startups. Fundraising time averaged 12 weeks for pre-seed rounds; the majority of successful seed companies closed in 12 weeks or less.
• DocSend, How to Create an Investor Strategy for Your Pre-Seed Fundraise — Russ Heddleston, 20 May 2021. Pre-seed report averages: 58 investors contacted, 30 meetings. If 30 real meetings all say no, change the pitch or make progress before contacting another 30; contact only investors actively investing, interested in your sector, and funding companies your size; more contacts bring more meetings, with a weaker link to amount raised.
• Paul Graham, How to Raise Money — September 2013. Be in fundraising mode or not; intros ranked; hear no till you hear yes; breadth-first search weighted by expected value; notes on scheduling less likely investors first (Jeff Byun) and asking for two intros (Joshua Reeves); ask what happens next; first substantial offer can be half the total difficulty, $50k from a well-known investor; close committed money ($800,000 example); avoid investors who do not lead; multiple plans; underestimate the target ($500k / $250k / $150k); accept offers greedily, 3 working days acceptable on an exploding offer; one founder handles fundraising; growth between meetings; send-the-deck-first as weak interest; stop when you get air in the straw; investors who reject you as warm leads.
• Y Combinator, A Guide to Seed Fundraising — Geoff Ralston. Raise to the next fundable milestone, usually 12 to 18 months; seed dilution up to 20%, avoid more than 25%; multiple plans; know your audience; warm introduction is by far the best way to meet investors; attempted close or clear next steps; close rapidly, SAFE signing in minutes; avoid investors asking for heavy diligence; part on good terms after a no.
• Gompers, Gornall, Kaplan and Strebulaev, How do venture capitalists make decisions? (Journal of Financial Economics) — Vol. 135, no. 1, January 2020, pp. 169–190. Survey of VCs, November 2015 to March 2016. Table 6: 73 days to close for early-stage firms vs 106 for late-stage.
• Carta, VC Fund Performance: Q1 2026 — Peter Walker and Kevin Dowd, 4 June 2026. 2,775 US venture funds closed 2017 through Q1 2026; about 89% smaller than $100 million; the majority of funds closed smaller than $25 million.
• Carta, As deployment rates slow, VC dry powder is piling up — Kevin Dowd, 12 September 2024. 2018, 2019, and 2020 vintages each deployed at least half of their capital in the first two years; 2022 vintage 43% deployed after two years.
• Carta, State of Pre-Seed: Q2 2026 — Hamza Shad, 13 August 2026. Few pre-seed deals exceed $2.5 million; those that do typically stack ten or more instruments.
• FounderNexus session — Operator judgment from live fundraising sessions: check what evidence investors expect behind the round label; start with a few relevant conversations before the investors who matter most; assign owners for customer work before outreach; “keep me in the loop” is a no; send short material-change notes with no ask to investors who passed. Not a survey.

Related

• Fundraising
• How seed VCs decide
• Seed pitch deck
• First angel check
• SAFE vs priced round
• Term sheet red flags
• Runway calculator with hiring plan

Founders who have closed seed rounds will pressure-test your milestone plan, fund list, and outreach order in a FounderNexus session before the first intro goes out.