What VC Application Forms Actually Screen For: How to Fill Them In and Pass the First Filter

VC application form tips only work if you understand who reads the form and how. At a programme receiving tens of thousands of applications a cycle, or a fund with a public Typeform, the first reader is an associate or a screening team with a rubric and a few minutes per entry, and their job is to find reasons to say no quickly. This guide, part of the Kurums Venture Capital hub, explains the anatomy of the forms, how triage actually works, the first-filter kill criteria, how to write the one-liner and traction fields, the numbers investors expect at pre-seed, seed and Series A, what the founder video must contain, how to follow up and reapply, and how to track a volume of applications without losing the thread.

What does a typical VC application form contain?

Six fields recur across every form: a one-sentence description, the problem and who has it, what you have built and what has happened since (traction), who the founders are and why them, the ask (round size, instrument, use of funds), and a link to a deck, demo or video.

The variation is in format and limits. YC’s current form, as documented by Causo Hub’s 2026 guide, caps the company description at 50 characters and the product answer at 50 words, asks for 100–150-word founder bios, a story about a non-computer system you hacked, a 50-word “why are you doing this”, the equity split and how long the founders have known each other, plus a one-minute video. Antler’s first form asks only for name, email, LinkedIn, stage and city, deferring everything else to interviews. Seedcamp’s FAQ lists qualities rather than fields. Speedinvest, Hustle Fund, Haatch and Fuel Ventures use 15–30-question Typeforms with a deck link. Public programmes go the other way: the EIC Accelerator’s short proposal is a 12-page form plus a 10-slide deck and a 3-minute video, and its full proposal runs to 20 pages with financials, letters of intent and a freedom-to-operate analysis. The full list of who runs which form is in the global guide to VC firms accepting open applications.

Because the fields repeat, the right preparation is a master document with the six answers in long and short versions — a 20-word, 50-word and 150-word one-liner; a 100-word and 300-word traction summary; a bio per founder in 50 and 150 words — that you then cut to each form’s limits. Founders who write each application from scratch introduce inconsistencies; founders who paste one master answer everywhere miss the fit question. The master document also becomes the first draft of the deck’s text layer, which keeps the two aligned.

How do associates actually triage applications?

In four passes, each faster than the last: an eligibility scan of a few seconds, a one-liner and traction read of a minute or two, a team-and-consistency check against the deck of a few minutes, and only then a call. Most applications are eliminated in the first two passes.

The eligibility pass is mechanical. Firms publish their rules precisely so that screeners can apply them without judgement: Seedcamp says companies that have raised more than a couple of million dollars are probably not a fit; HTGF requires a German base and a company under three years old; Blume says no idea-stage deals for first-time founders; YC requires a US, Canadian, Cayman or Singapore entity before investing; SEIS funds require SEIS qualification. An application that fails one of these is closed before the product is read, and the founder usually never learns why.

The second pass is the one-liner and the traction field, read together. The screener wants to know in one sentence what the company sells, to whom, and why it wins, and then wants one number that proves someone cares. DocSend’s research on how investors read decks found that they now spend under three minutes on a deck, and that successful pre-seed decks held attention for over four minutes against about a minute and a half for unsuccessful ones; the same asymmetry applies to forms. If the one-liner forces the reader to work, or the traction field is a paragraph of adjectives, the entry is closed.

The third pass is where pattern-matching happens. The screener opens the deck, checks that the traction number on the slide matches the one in the form, reads the team slide against the bios, and looks for the things that signal risk — a part-time founder, a dead-equity co-founder, a round size that does not match the stage. The founder red flags that end fundraising are exactly what this pass is looking for. Only applications that survive all three reach an associate call, and only a fraction of those reach a partner. The funnel is brutal but it is also predictable, which is what makes it beatable.

What are the first-filter kill criteria?

Wrong geography or entity, wrong stage, wrong sector, a one-liner without a customer, no traction number, internal contradictions, and an ask that does not match the firm’s cheque size. Each is published or inferable from the firm’s page, and each is fatal on its own.

Published rules are taken from the firms’ own pages as reviewed in October 2026; screening norms reflect consistent practice across programmes and funds.

The ask deserves special attention because it is the easiest kill to avoid. Every firm with an open route publishes a cheque range, and the application that asks for the wrong amount is read as evidence that the founder has not read the page. Equally, a round that is too small for the stage (a $200k seed for a company with $1M ARR) or too large (a $5M pre-seed for a prototype) signals a misunderstanding of the market. The fundraising process and timeline guide explains how to size a round, and the barbell effect analysis explains why 2026 round sizes are polarised between seed and mega-rounds.

How do you write a one-liner that survives triage?

Name the customer, the product and the reason you win, in that order, in under twenty words, with no adjectives: “Dispatch software for mid-size trucking fleets that cuts empty miles 15% using live load data.”

The formula is customer, product, mechanism, outcome. “Mid-size trucking fleets” tells the screener who pays and how big the market might be. “Dispatch software” tells them the category and the comparables. “Using live load data” tells them the mechanism and hints at defensibility. “Cuts empty miles 15%” gives a number they can test. Compare the entries that fail: “AI-powered logistics platform” names no customer and no mechanism; “The Uber for freight” borrows a comparison instead of making a claim; “Revolutionising how fleets move goods” is a slogan.

Three tests catch most weak one-liners. Could a stranger repeat it accurately after one reading? Does it contain a noun a buyer would use to search for you? Would a competitor dispute it? If the answer to the third is no, the line is too generic. Write ten versions, read them aloud, and pick the one that passes all three — then use it unchanged everywhere: the form, the deck’s title slide, your email subject lines, your LinkedIn headline. Consistency across surfaces is itself a signal that the founders know what they are building.

Which numbers matter at each stage?

Pre-seed is judged on evidence of demand and speed of shipping; seed on revenue or usage with growth and retention; Series A on ARR, net revenue retention, gross margin, CAC payback and burn multiple. Every number needs a definition, a period and a trend.

Thresholds vary by firm and sector; Blume, for example, publishes revenue, GMV and MAU guidelines on its site. The pattern — a defined number, a period, a trend — does not vary.

Two rules apply across the table. First, define every metric. “Revenue” can mean bookings, billings, GMV, gross or net; “retention” can be logo, revenue or cohort; “growth” can be monthly, quarterly or year-on-year. The screener who has to guess will assume the least flattering definition. Second, give the trend, not just the level. $38k MRR means little; $38k MRR that was $12k five months ago means a great deal. If a number is weak, say so and say why; a transparent weak number survives triage better than a hidden one, because the third pass will find it anyway. The investor-ready pitch deck and data room guide sets out the metrics per slide and the data-room documents that back them, and the data on which founders get funded shows how team signals weigh against traction at each stage.

How should you handle the founder video?

Keep it under the limit, put every founder on camera, speak directly to the lens, and say four things each: who you are, what you are building in one sentence, why you, and one concrete recent event. No slides, no music, no editing unless the programme allows it.

YC’s rule is explicit: one minute, all founders on camera, nothing but the founders talking, with a recorded video call acceptable. The EIC asks for a three-minute pitch video with its short proposal. Most Typeform-based funds make the video optional, and an optional video that is well made is a differentiator because so few founders bother. In all cases the screener is testing three things the text cannot show: whether the founders agree on what the company is, who leads and whether the others defer or contribute, and whether the energy is real.

Practical rules: record in daylight with a plain background, use the laptop camera at eye level, do not read from a script but do rehearse until the four points come out in order, and have each founder speak — the quiet co-founder who says nothing raises a question the form cannot answer. Watch it back once with the sound off and ask whether these are people you would back. If the answer is no, re-record. For programmes with a live interview, the video is also the rehearsal: YC’s ten-minute interview, Antler’s problem-solving sessions and Techstars’ managing-director calls all reward the same directness.

What do strong and weak answers look like side by side?

Strong answers are specific, numeric, linked and short; weak answers are adjectival, unbounded and unverifiable. The difference is visible in every field, and screeners are trained to see it.

Illustrative entries written by the Kurums editorial team; company details are hypothetical.

How should you follow up, and when should you reapply?

Send one follow-up after the firm’s stated response window, or after two to three weeks if none is stated, containing one new fact. Reapply when something material has changed, and say what changed in the first sentence. Never send a second application that is weaker than the first.

Only a few firms publish timelines — 1517 Fund says about two weeks for in-thesis submissions, Seedcamp aims for two weeks from first meeting to decision, YC gives a decision date for on-time applicants, the EIC gives four to six weeks for short proposals — and Speedinvest states that only shortlisted founders are contacted. Silence is therefore the default outcome, and the follow-up is the only tool a founder has against it. It should be three sentences: what you applied for and when, one new fact since then (a customer, a number, a hire), and a one-line ask for a reply. One follow-up is professional; three are noise.

Reapplication is encouraged more than founders assume. YC states that many accepted companies applied more than once, and with four batches a year a founder can reapply within three months. Techstars programmes, Antler residencies and most Typeform-based funds keep prior submissions and read the new one against the old. That makes the reapplication a progress report: open with what changed, quantify it, and keep everything else consistent. A reapplication that shows a worse number than the original, or that reinvents the one-liner without explanation, reads as drift and is closed quickly. The YC, Techstars and Antler guide covers each programme’s reapplication norms, and the warm introduction guide explains how to convert a cold rejection into a referred second look through a portfolio founder.

How do you track applications at scale?

Keep one tracker with a row per firm or programme and columns for window dates, cheque, equity, eligibility notes, the link to your submitted answers, submission date, reply date, status and next action; review it weekly, and never let two live applications carry different numbers.

A spreadsheet works for 30–50 targets; beyond that a fundraising CRM is cheaper than the mistakes. The columns that matter most are the eligibility notes (so you do not apply where you cannot win), the link to the exact answers you submitted (so a follow-up or interview does not contradict them), and the next-action date (so follow-ups happen once, on time). Tag each row with the route — programme, open form, cold email, warm intro — so you can see conversion by route after a month; most founders find that warm introductions convert several times better than forms, which is a reason to run both rather than a reason to abandon forms. Our comparison of fundraising and investor CRM software covers the tools, and the pitch deck software comparison covers the deck tools that produce the trackable links screeners expect.

Finally, treat each cycle as data. Record which one-liner version was live when a firm replied, which traction number was in the form, and what the first question on any call was. After twenty applications the pattern is usually obvious — the one-liner that gets replies, the number that gets questions, the filter you keep failing — and the next twenty go better. The regional lists of who to apply to are in the Europe and UK guide and the Asia, Middle East and Turkey guide, and the guide to what a strong pitch contains applies to the call that follows a successful form. Browse the full VCs Accepting Applications pillar for the companion guides.

Frequently Asked Questions