Can You Raise Funds with Just a Validated Idea? The €200 Test Investors Trust

A founder we'll call M. spent six weeks rewriting his pitch deck. Sequoia template, YC structure, fourteen versions of the executive summary. He raised €0.

That same week, another founder ran €214 of Reddit ads at a one-page Webflow site with a "reserve your spot" button. He came back with 412 emails, a 4.2% landing-page conversion rate, and three angel cheques inside the next three weeks.

That's not luck. That's the difference between selling an idea and selling signal.

Investors don't fund ideas. They fund signal. The cheapest way to manufacture signal in 2026 is a €200 paid-traffic test pointed at a landing page — and the data it produces carries more weight in a pitch room than the next ten pages of your deck.

This article is the playbook. The mechanics, the numbers, the slide-ready output, and the cases where it's the wrong move. We've watched dozens of founders run this exact test through LemonPage; the patterns below are the ones that hold up.

What investors actually do with your deck

Start with the most counter-intuitive number we know.

DocSend tracks pitch-deck behaviour across thousands of seed rounds. Their finding: the average investor spends 2 minutes 24 seconds on a deck. Seed decks specifically average 3 minutes 20 seconds. 31% of seed decks bounce within 10 seconds. (DocSend Pitch Deck Metrics)

You wrote that deck for fourteen weekends. They will read it for the time it takes to drink half an espresso.

But here's the twist that changes the strategy. DocSend's same dataset shows that investors spend 80% more time on the traction slide of decks that don't raise. They're not impressed; they're trying to figure out why the numbers don't add up. A messy traction slide is worse than no traction slide — it gives the room something to chew on, and what they chew on is your weakness.

The mandate is clear. Either show traction that survives a 90-second skim or show none at all and lean entirely on team and market. The middle option — soft, hand-wavy traction — is the one that sinks rounds. We see it every week.

Why a €200 test outperforms a 30-page deck

Three things make a paid-ad smoke test the highest-leverage signal you can manufacture before raising:

• It's adversarial. Strangers paid by Meta or Reddit to see your offer click only if the offer is for them. Your mum, your roommate, your accelerator cohort — all polite, all useless. Paid traffic doesn't care about your feelings.
• It's reproducible. A VC can re-run your numbers. Ad-platform screenshots are auditable. A user-interview deck is a story; a Meta dashboard is evidence.
• It's cheap enough to repeat. €200 buys roughly 100–300 clicks on Reddit, 50–150 on Meta in a B2B niche, depending on how saturated the keyword is. Two iterations cost less than a Saturday-Sunday hackathon.

Compare that to the deck. The deck is a story you tell yourself before strangers can disagree with you. The test is strangers disagreeing with you.

The pre-seed reality: most teams don't have revenue

The good news for founders without product: less than 20% of pre-seed teams had paying customers in 2025, down from 31% in 2024. (This stat is reported by Fusion VC citing Carta — we couldn't fully verify it inside Carta's published summaries; treat as directional.)

What we know for sure from Carta's State of Pre-Seed 2025: pre-seed startups raised $4 billion across 25,000+ convertible instruments in 2024. Median round size hovers around $700K. SAFEs are used in 92% of those rounds. 35% of Q4 2025 rounds were under $250K — a new high in micro-checks.

So at the bottom end of the market, founders are raising on something other than revenue. Often it's relationships. Sometimes it's reputation. The honest gap in pre-seed advice is: nobody tells you what to show if you have neither.

A €200 paid-traffic test is the answer. Specifically.

The €200 test mechanics

Here's the playbook. We've run this dozens of times; the steps below are the ones that survive contact with reality.

Step 1 — One-page pitch site, 15-20 minutes of work.
Not a polished marketing site. One page. Headline, sub-headline, the offer, a single CTA. Use Webflow, Framer, Carrd — or validate the idea in LemonPage if you'd rather skip the page-builder learning curve. Headline must be the promise of the result, not a feature list. "Stop paying €10 a trade" beat "commission-free brokerage" for Robinhood. (Prefinery teardown)

Step 2 — A real CTA. Not "join the waitlist".
Waitlist signups are cheap. They feel like signal but they're closer to social politeness. Three CTAs that produce real signal:

• "Reserve your spot for €5 (refundable)." Best for B2C. The €5 buys you near-perfect kill-criterion data: people who don't pay aren't your buyers.
• "Pre-order at 50% off launch price." Best for indie SaaS. Stripe Payment Link, takes 4 minutes to set up.
• "Book a 20-min onboarding call." Best for B2B. Conversion is lower but quality of signal is much higher.

If you absolutely must use a waitlist, add a refundable deposit so the conversion isn't theatre.

Step 3 — €200 of paid traffic, single channel.
Don't split between Meta and Reddit and Google. Pick one. The ad benchmarks for 2026 (sources at the bottom of this article):

• Reddit: B2B SaaS CPC $0.50–$2.00. Best for indie SaaS, dev tools, niche communities. CPCs are 40-80% cheaper than LinkedIn.
• Meta: Median CPM $13.48 across all industries. B2B SaaS specifically: CPC $10.23, CTR ~1.1%, CVR ~2.3%, CPA $55. Pricier per click; better for B2C and visual products.
• Google Search: B2B SaaS non-brand CPC $8.50–$14.00 in 2026, climbing. Use only if your audience search-shops for the problem you're solving (so: dev tools, productivity SaaS, niche utilities).

Run a single ad set against the cheapest plausible channel for 7 days. Spend €30/day. Stop on day 7 and look at the data.

Step 4 — The five numbers that matter.
After the run, you should have:

1. Click-through rate (CTR) — out of 100 ad impressions, how many clicked.
2. Landing-page conversion rate (CVR) — out of 100 visitors, how many took the CTA.
3. Cost per lead (CPL) — €200 ÷ leads. The single number a VC will ask for first.
4. A qualitative comment thread — Reddit ad comments, Meta engagement, replies. The VC will ask "what did people say?"
5. A breakdown of who clicked — geography, device, demographic. Deck-grade colour.

That's five numbers and one paragraph of qualitative. It fits on one slide.

What "good" looks like

Here are the thresholds we've watched separate "the test worked" from "the test bombed":

The Unbounce 2024 dataset (41,000 SaaS pages, 464M visitors) puts SaaS landing-page conversion median at 3.8%, top quartile at 11.6%+. (Unbounce SaaS report) That's the real anchor. Anything above 3.8% is genuinely above-average. Anything below 2% means the offer or the audience is wrong, and no amount of deck-polishing will fix it.

A small note worth absorbing: landing pages written at 5th-7th grade reading level convert at 12.9% vs. 2.1% for "professional-level" copy. Same data set. Founders who write like McKinsey associates lose; founders who write like they're explaining the product to their cousin win.

Translating the test into deck slides

The traction slide is the only one DocSend says investors meaningfully scrutinise. Here's the structure that works:

• Slide title: A flat factual claim. "412 leads from €214 in 7 days. €0.52 per lead. 4.2% landing CVR." Not "Strong early demand". Numbers, not adjectives.
• One screenshot. The Meta or Reddit dashboard, dated. Investors who care will check; investors who don't will skim and see "ah, real numbers."
• The two qualitative quotes. Pull two replies, comments, or DMs from real prospects. Make sure one is sceptical-but-curious, not just enthusiastic. Investors trust mixed signal more than uniform praise.
• The kill-criterion line. Something like: "We pre-committed to abandon if CVR < 2%. We hit 4.2%, so we're proceeding." This is the line we see VCs respond to most. It signals you didn't hill-climb to a positive interpretation; you set a threshold and beat it.

That's one slide. The rest of the deck is team, market, mechanism, ask. The traction slide is the only one that gets re-read.

When the €200 test isn't enough

Five situations where the playbook above fails or misleads:

• Deep-tech. Amadeus Capital is explicit: pre-seed deep-tech investors evaluate "technical risk first, market risk second, execution risk third." A landing-page test for a quantum-photonics chip says nothing useful. Validation evidence here is the technical milestone — a working prototype, a benchmark beat, a published result.
• Regulated industries. Robinhood famously needed VC capital for regulatory approval. They couldn't get the approval without the money. A landing-page CVR doesn't unlock a banking licence; it just signals demand exists, which everyone already assumes.
• Network-effect plays. Marketplaces, social products, multi-sided platforms. A landing-page test flattens them into a single offer; the actual product value emerges only after several interactions. Use an MVP-first approach here. Try to validate one side first (the harder side, usually supply).
• Enterprise sales >$50K ACV. A 20-minute onboarding call with three CIOs is more credible than a Reddit-ad CTR. The signal at this end of the market is named-logo conversations, not paid traffic.
• The murky middle. This one's the trap. If your test produces a 2.4% CVR and €4 CPL, you don't have validation; you have ambiguity. Investors spend 80% more time on traction slides of decks that don't raise. Murky-middle traction is exactly what they latch onto. If the test bombs cleanly, kill the idea or pivot the offer. If it lands cleanly, raise. If it's middle, run another test before stepping into a pitch room.

The vanity-waitlist trap

A specific pitfall worth its own section.

Brian Nichols of Hustle Fund — one of the most active pre-seed investors in the U.S. — has been public about it: "A list of 500 people with a 40% conversion rate is infinitely more valuable than 10,000 names with a 2% conversion rate."

This is the trap dozens of founders walk into. They run a waitlist sweepstake (refer-a-friend, viral mechanic), accumulate 8,000 emails, and walk into the pitch room with "we have a waitlist of 8,000". The first sharp investor asks: what's the post-launch conversion to paying customer? If the answer is "we don't know yet" or "we're tracking 1.5%", the founder just demonstrated that 7,880 of those emails are noise.

Robinhood's 1M-person referral waitlist is the canonical case for and against this approach. It produced enormous launch-day buzz. It also — and founders rarely tell this part — was rejected by 75 VCs before the first cheque. The waitlist was insufficient. The eventual round was a function of the founders' fintech credibility, market timing post-2008, and a thesis-driven seed investor (Index) finally biting.

So if you're running a waitlist test, instrument the conversion rate. Don't lead with the headcount.

What three real cases look like

A useful set of reference points, ordered from cleanest to messiest:

Buffer (2010). Joel Gascoigne ran a two-page landing site implying Buffer existed, with a fake "Plans & Pricing" button leading to a "we're not ready, leave your email" page. Result over 7 weeks: 120 email signups. 50 of those 120 became users on launch day. First paying customer 3 days later. Joel's own framing: "I didn't get ‘a billion signups’, in fact in a long 7 week period I only got 120 signups. But I spoke with a lot of those people during that time." Buffer self-funded into product-market fit. The smoke-test was the origin, not just an experiment.

Dropbox (2007). Drew Houston shipped a 4-minute screencast demo of a product that didn't exist, posted it to Hacker News, and pinned the YC application to its reception. Pre-video waitlist: ~5,000. Post-video, 24 hours later: 75,000. Sequoia led a $1.2M seed in 2007. Drew, in his own words on X this past March: "My (perhaps questionable) strategy for getting into @ycombinator: 1) Post a Dropbox demo on Hacker News 2) Pray @paulg @jesslivingston see it. It actually worked." Different mechanics from Buffer (no paid ads), same lesson — the test produced a number Sequoia could verify.

Scoutnow.ai (2024-2025). AI-recruiter MVP gated behind a waitlist; one free resume scan unlocked. They packaged a 42% activation rate and 25% referral rate into the pre-seed deck and closed $750K in 11 days. (Single-source reporting; treat as illustrative, not verified.) The pattern is the same — a small validation experiment, two numbers worth showing, a deck built around them.

The structural lesson across all three: the validation experiment becomes the deck. Not a bullet point inside the deck — the spine. Buffer's 120 signups, Dropbox's 75K waitlist, Scoutnow's 42% activation. One number per company, decisive enough to base a slide on.

Validation shortens the conversation. It doesn't replace it.

We've spent 2,400 words arguing that a €200 test outperforms a deck. The honest finish: it's necessary, not sufficient.

Andrew Chen's framing from a few years back still holds. "Pre-seed: bet on the team / Seed: bet on the product / Series A: bet on the traction / Series B: bet on the revenue." (andrewchen.com) At pre-seed, validation data complements founder credibility — it doesn't substitute for it. A weak founder with a clean €200 test still loses. A strong founder with no signal still raises sometimes. The €200 test is the cheapest possible upgrade to your founder pitch, not a magic key.

What it does do — and this is the unsexy real value — is shorten the room. Investors who would have asked twelve clarifying questions about market size now ask one. Investors who would have wanted three follow-up calls book the second meeting on the spot. The signal isn't the cheque; it's the slope of the conversation.

If you've been rewriting the same deck for six weeks, you have your answer. Spend €200 on traffic this Saturday. Walk into next week's investor meetings with one slide of real numbers. The deck can stay broken — the slide just has to work.

Validate your idea on LemonPage in under an hour. We built it because we kept losing 4 hours of plumbing per test, and the math of validation-as-fundraising-tool only holds if running tests is cheap.

FAQ

Can you raise pre-seed funding with just an idea, no product?

Yes — under 20% of pre-seed rounds in 2025 had paying customers. But "just an idea" doesn't fly. You need some externally verifiable signal: a paid-ad smoke test, a demo video that drove a real waitlist, technical milestones for deep-tech, or a track record. Idea-only raises are reserved for repeat founders investors already know.

How much should a validation test for fundraising cost?

€100 to €300 is the standard range. Below €100 you don't get statistically usable numbers; above €300 you're in iteration territory, which is fine but no longer a "first test". Think of the budget as a quality bar — €200 is the line at which the data is decisive enough to base a deck slide on.

What conversion rate signals real demand to investors?

Above the 3.8% SaaS median (Unbounce 2024 dataset, 41K pages). Above 8% is genuinely strong. Below 2% is bombed. The middle 2-3.8% range is the murky zone — investors will scrutinise it harder, not less.

Will a waitlist of 10,000 people get me funded?

Not on its own. Hustle Fund's Brian Nichols is on record that a 500-person list with 40% conversion beats a 10,000-person list with 2%. The number that matters is downstream activation, not headcount. Always lead with the conversion rate, never the list size.

What's the difference between a €200 test and a real MVP?

The test answers "is there demand?" with a CVR. The MVP answers "does the product solve the problem?" with usage and retention. They're sequential, not interchangeable. Run the test first. If it bombs, don't build. If it lands, build the smallest MVP that lets you measure retention. (More on the order of operations.)

Does a paid-ad test work for B2B with high contract values?

It produces weaker signal at high ACV. For €50K+ B2B, a 20-minute call with three CIOs from your target ICP carries more weight in the deck than a Meta dashboard. Use the smoke test to source the calls; use the calls as the actual evidence.