Warm Intros Are Costing Founders Real Money
Every fundraising guide tells you the same thing. Build your network. Get warm intros. Never cold email a VC.
One founder sent 400 cold emails. He built his entire pre-seed round from them.
The conventional wisdom on seed fundraising is outdated - and founders who follow it are losing months of runway chasing introductions that never come.
This guide covers how seed fundraising works right now. Theory is not going to move your round forward. Real mechanics and real numbers are what close deals in the current market.
What Seed Funding Is (And What It Isn't)
Seed funding is the first substantial outside investment in your company. It comes after you have something to show - a prototype, early customers, or at minimum a very clear problem and a credible team to solve it.
Seed funding is for companies that have something to show. Pre-seed is where you take the raw idea and figure out if it deserves to exist.
The purpose of seed is to take you from early validation to the point where a Series A investor can make a data-driven decision. That means product-market fit signals, consistent growth, and unit economics that don't make investors wince.
The Current Numbers
Here is what the seed market looks like right now, based on Carta, PitchBook, and CB Insights data:
- Median U.S. seed round size: $2.5M - $3.5M, with $3.1M as the central benchmark (Carta)
- Median pre-money valuation: $14M - $20M depending on sector and traction
- Typical founder dilution at seed: 15% - 20%
- SAFEs used in 92% of pre-seed rounds as of Q3 (Carta)
- AI startups command a 1.3x - 1.6x valuation premium over non-AI comparables (CB Insights)
- 66% of top-decile seed valuations go to Bay Area (44%) or New York (22%) startups (Carta)
Those numbers look clean on paper. The reality is messier. Fewer startups are getting funded at all, but the ones that do are raising at higher valuations. The median went up. The volume went down. Founders conflate the rising valuations with rising odds of getting funded.
AI-Generated Pitch Decks Are Now a Liability
AI-generated pitch decks are now a liability.
A venture investor who runs their own fund posted recently that the quality of fundraising decks has dropped substantially - and flagged AI offloading of critical thinking as a likely cause. The post punched well above its weight in engagement for the account size. It hit a nerve because investors are seeing it every day.
Every deck looks the same now. Same slide order. Same vague TAM methodology. Same "10x better" positioning with no proof. When your deck looks like it was assembled in 20 minutes from a GPT prompt, it signals one thing to the investor: this founder hasn't done the hard thinking yet.
The founders closing rounds right now are doing the opposite. They are showing radical transparency. One founder posted the exact deck that raised a $9M seed round and got nearly 200,000 views and over 1,600 likes because he said "copy whatever you want." The engagement wasn't about the money. It was about the willingness to show the actual work.
A pitch deck is proof that you have thought harder about this problem than anyone else in the room.
What VCs Are Funding Right Now
Ask a VC what they care about and you'll get a polished answer about market size and founder-market fit. Ask a founder who just closed a round and you'll get the list.
From Reddit threads with practitioners who have multi-year VC backgrounds, combined with the highest-signal content from active investors posting publicly, the priority order looks like this:
Find Your Next Customers
Search millions of B2B contacts by title, industry, and location. Export to CSV in one click.
Try ScraperCity Free1. Team and Founder Conviction - Before Everything Else
At seed, the product is still a work in progress. The market thesis might be wrong. The go-to-market will definitely change. The team has to hold through all of that.
"Founder-market fit must be exceptional" is how one Southeast VC phrased it - the founder who knows this problem from the inside, not the generalist who read about the opportunity.
Investors are pattern-matching on whether you will survive the 24 months between now and Series A. They are betting on you more than your deck.
2. Market Size - Bigger Than You Think You Need
VCs need a fund-returning outcome to make their math work. That means they need the company to potentially be worth $1B+. So the market needs to be enormous - hundreds of billions minimum for most institutional funds.
The target return math is simple: a seed VC investing $2M needs 20-30x to make the fund work. That is $40M - $60M back from one check. That only happens if the company gets big.
If your market is $500M, you are a good business. You are not a venture business. Know the difference before you walk in.
3. Defensibility - The Slide That Closes Deals
One tweet with 685 likes put it this way: if you have a real moat, throw away the entire deck and just make one slide on that.
Defensibility separates fundable from unfundable at seed. What happens when a well-funded competitor copies you? Why does being first to market matter? What makes your position harder to attack six months from now than it is today?
If you cannot answer that clearly, investors will assume the answer is "nothing."
4. Traction - The Number That Changes Everything
YC uses 10% week-over-week growth as the threshold that makes investors take notice. That is not a casual benchmark. At 10% week-over-week, you move from interesting to fundable in the YC model.
You don't need to be at that number to raise. But the closer you are, and the more clearly you can show the trajectory, the more leverage you have in every conversation.
One accounting startup - in one of the slowest-moving industries in B2B - closed $520,000 in revenue in 60 days with two people. No inbound. No reputation. Just a targeted outbound system hitting the biggest financial pain points with hyper-personalized offers. That traction story raised $3M in seed funding and positioned the company for a $10M+ Series A.
That is what traction looks like to a seed investor. Not a waitlist. Not 500 Twitter followers. Revenue, growth rate, and a story about why the next 90 days look like the last 90 days.
5. Unit Economics - Show the Math
LTV needs to beat CAC. The business needs to get more efficient as it scales, not less. Payback period needs an explanation.
Investors are not expecting perfection. They are expecting that you have thought about this clearly. A founder who can walk through their unit economics from memory - including where the model breaks and why it doesn't break in their case - closes faster than one who has to find the slide.
The Lean Team Signal
Headcount is now a signal, not just a cost line.
According to Carta data, average seed-stage headcount has dropped around 40% - from roughly 10 employees at seed in the early part of this decade to about 6 today. Walking into a pitch with a bloated team is now a red flag. It suggests poor capital efficiency and unclear role definitions.
The founders raising the best rounds right now are doing more with fewer people. Two or three sharp operators with a clear division of labor beats eight generalists every time in the current environment.
Want 1-on-1 Marketing Guidance?
Work directly with operators who have built and sold multiple businesses.
Learn About Galadon GoldIf you have a 10-person team and have not raised a dollar yet, be ready to explain why.
How Much to Raise
I see this every week - founders raising too little and running out of runway right before a milestone, or raising too much and setting valuation expectations they cannot meet at Series A.
The right number is calculated backward from your milestones, not forward from what you think you can get.
A simple framework: what do you need to achieve to make Series A a clear yes? Work backward from that milestone. Calculate your burn rate to get there. Add a 30% buffer for the unexpected. That is your raise target.
Engineering costs roughly $15,000 per month per person all-in. Multiply by headcount, multiply by months needed, and you have a defensible number to put in front of investors.
Target 18 months of runway minimum. Many advisors say 24 months. The reason is simple: seed rounds take 3 - 6 months to close. If you start fundraising with 6 months of cash left, you are negotiating with a gun to your head, and investors can smell it.
Start conversations 6 - 9 months before you need the money. Use the first months to build relationships. Intensify when you have 4 - 6 months left.
SAFEs vs. Priced Rounds - What to Use
SAFEs have become the dominant instrument at pre-seed and early seed. According to Carta, they are used in 92% of pre-seed rounds. They are faster to close, cheaper to execute legally, and let you raise without setting a hard valuation.
A SAFE works like this: you agree on a valuation cap - the maximum price at which the investor's money converts to equity when you do a priced round later. The investor is not getting shares today. They are getting the right to buy shares at a discount or capped price in the future.
For rounds under $1M, $10M is the current market median cap. For rounds between $1M - $2.5M, $15M is typical, per Carta's current data.
The trap: SAFEs stack quietly. If you raise three SAFE rounds at different caps before your Series A, all of those convert at once. Founders who don't model this regularly discover they have given away far more than they realized. The rule is to model your full cap table before signing every single new SAFE.
One critical note: use post-money SAFEs, not pre-money SAFEs. Pre-money SAFEs create ownership uncertainty when you stack multiple instruments because each new SAFE recalculates the others' ownership percentages. Post-money SAFEs are now the market standard for a reason.
Once your deal crosses $4M - $5M, priced rounds start to dominate. Over 70% of seed deals above $5M are structured as equity financings. If you are raising a larger round, a priced round signals that you are ready for institutional governance - which can make the next check easier to close.
The Dilution Math You Need to Know
The median founder gives up 19.5% in their seed round, per Carta's dataset. Add a 10% option pool top-up and total ownership drops from roughly 80% to around 60% after seed.
Stack a Series A on top of that - another 20% dilution - and you are down to around 36% after two rounds. Add Series B and you are at 23%. That is before any bridge rounds or additional dilution events.
None of that is bad if you are building something large. But I've watched founders give up 25% at seed, then 25% at Series A, then express shock at owning 15% at Series B - they didn't run the math early enough.
Find Your Next Customers
Search millions of B2B contacts by title, industry, and location. Export to CSV in one click.
Try ScraperCity FreeThe target: keep seed dilution under 20%. Under 18% if you can. That buys you room to stay in the driver's seat through Series B and beyond.
How to Find Investors Who Are Writing Checks
I see this every week - founders building a list of big-name VCs, blasting out cold emails, and wondering why nothing converts.
The problem is fund cycle timing. Only about 10% of venture funds are actively deploying capital at any given point. If you email a firm that closed their last fund three years ago and hasn't started a new one, that email is going nowhere regardless of how good your company is.
Filter for investors who have made at least one investment in the past 3 - 6 months. That is the baseline check for active deployment.
Second filter: match your round size to the fund size. Investors write checks at roughly 1% - 2% of their total fund size. A $50M fund writes $500K - $1M checks. A $200M fund writes $2M - $4M checks. If your round needs a $3M lead and you are emailing a $30M fund, you are wasting both your time and theirs.
Third filter: stage specialists beat stage tourists every time. A fund that leads seed rounds as their core strategy will move faster, add more value, and be easier to work with than a growth-stage fund dipping into seed opportunistically.
The Cold Email Playbook That Works at Pre-Seed
The conventional wisdom says cold email doesn't work with VCs. The data says otherwise - if you do it right.
One founder documented sending just over 400 cold emails that formed the backbone of their entire pre-seed raise. Cold outreach, personalized correctly.
What makes the difference between a cold email that gets a reply and one that gets ignored?
Personalization. Traction. The third-party pitch.
Personalization that proves you did homework. Specific. Why this investor? What did they write about that connects to your thesis? Does your startup complement or contrast with a portfolio company of theirs? Two sentences of real homework beats two paragraphs of flattery.
Traction that changes the math. A cold email from a founder with $50K MRR and 40% month-over-month growth is not the same as a cold email from a founder with a deck and a dream. If you have traction, lead with it. That is the sentence that determines whether the next sentence gets read.
The third-party pitch. The highest-performing piece of content on warm intros vs. cold email - over 550 likes in our analysis - made a simple argument: hiring an advisor to cold email VCs on your behalf outperforms founder-direct cold email. The third-party pitch gets past the spam filter because it signals that someone credible has already vetted you.
If you have an advisor with a track record, have them make the introduction. If you don't, work on getting one before you start the process.
The Forwardable Blurb - The Most Underused Tool in Fundraising
I see this every week - founders who get a warm intro and blow it.
They ask a connection to "put them in touch" with an investor. The connection says sure. The connection sends a vague email saying "you should meet my friend." The investor says thanks and moves on.
Give your champion a pre-written referral blurb they can forward verbatim. A referral script.
It needs five things: who you are in one sentence, what specific problem you are solving and for whom, what traction or proof you have, why this investor specifically would care, and a clear ask - a 20-minute call, not a request for funding.
The goal is to make it effortless for your champion to advocate for you. They should be able to forward your blurb without editing a word. Every edit they have to make is extra work that kills the chance of it happening.
Write the email for them. All they have to do is hit send.
Your Pitch Deck - What Goes In, What Comes Out
A seed pitch deck is a story that ends with the investor believing two things: this market is huge, and this team is the one to win it.
The slides that matter most, in rough priority order:
The problem. Make it visceral. If the investor doesn't feel the pain in this slide, nothing else matters. Lead with customer quotes, not market statistics.
The solution. One sentence. If you need three sentences to explain what your product does, you don't understand it yet.
Traction. The single most important slide in the current market. Revenue, growth rate, retention, NPS - whatever you have. Show the direction. Show the acceleration if it's there.
Market size. Show TAM, SAM, and SOM - but do not just cite a Gartner report. Show how you calculated it from the bottom up. Investors have seen the inflated top-down numbers a thousand times.
The moat. Per the 685-like tweet: if you have a real defensibility story, this is your whole deck. What do you have that a well-funded competitor cannot copy in six months?
Team. Why are you and your cofounders uniquely positioned to win this specific problem? Not "experienced operators" - specific, relevant experience that makes this team the right team for this exact problem.
The ask. How much are you raising, at what terms, and what milestones will it fund? Be specific. "Raising $2.5M to reach $500K ARR and hire three engineers" is better than "raising a seed round to grow the business."
What to cut: slides on your technology architecture unless it is the defensibility story, three-year financial projections that have no basis in reality, and any slide that exists to make you feel like you covered all the bases.
Running the Process - What Closes Rounds
Fundraising is a sales process. Treat it like one.
The mechanics that matter:
Run parallel, not sequential. Don't take one meeting, wait for feedback, then take the next. Have 20 conversations in the same two-week window. This compresses your timeline from 6 months to 6 weeks, surfaces consistent objections you can address, and creates urgency.
Get to no fast. Time spent chasing a soft yes from the wrong investor is time not spent finding the right one. Ask directly: "Based on what you've seen so far, is this the type of deal your fund does?" A fast no is worth more than a slow maybe.
Oversubscribed beats barely funded. One founder with a warm network closed a seed round and a Series A in about three weeks each once they had traction. The difference between that and 18 months of misery was not the pitch deck. It was traction and timing.
Start with angels. Angels move faster than institutional VCs, take smaller checks, and create social proof. Getting your first $200K from three angels who have strong reputations often makes the next $1.8M from a lead VC significantly easier.
One founder who sent 432 cold emails, did 120+ pitches, and hit what he called the "valley of death" at less than 50% of his target - still closed an oversubscribed round. Persistence is not a soft skill in fundraising. It is the primary skill.
The Rejection Is the Process
Over 12% of relevant content in our analysis of seed fundraising discussion was explicitly about rejection - getting passed on, hearing no, being ghosted. The data shows you what the process looks like.
In my experience, no's are about fund stage, portfolio conflicts, check size mismatch, or the investor already having a bet in your category. None of those things are feedback on your idea.
The founders who raise collect no's faster than they collect yes's. The ratio is just that bad for everyone. The ones who don't raise are the ones who slow down after the first ten rejections and spend time trying to figure out what went wrong instead of sending the next ten emails.
One operator who went through a seed raise described it this way: the goal is not to avoid rejection. The goal is to identify which investors are worth pursuing and which ones are polite time-wasters - and to make that determination in a single conversation rather than across six months of follow-ups.
The Accelerator Question
Should you apply to YC or another top accelerator before raising?
The answer depends on where you are.
YC gets roughly 23,000 applications per batch with under 1% acceptance. The Spring batch recent data shows 35% of companies rank in the top 20% of all YC startups ever, and 14 companies were already past $1M ARR at entry. That is the competitive set you are entering.
If you get in, the math changes dramatically. Strong YC companies raise $5M - $20M seed rounds within weeks of Demo Day. The network effects compound and the signal opens doors that are otherwise closed.
Rejection is a signal to fix something before your next application or to raise outside the YC track entirely.
The mistake is using "I'll apply to YC first" as a delay tactic. If you have traction and a working product, you can raise without an accelerator. Waiting 6 months for a batch slot when you have momentum costs you exactly that momentum.
The Numbers Nobody Gives You
Here are the figures most seed fundraising guides skip over:
- Median seed cash raised: $4.0M (Carta). The median post-money valuation for a seed round is $20.0M.
- AI premium: AI software companies command roughly $19M median valuations vs. $15M for the broader market (Carta).
- AI's share of seed capital: AI is capturing 41.7% of all seed capital right now (Carta).
- Geographic premium: 66% of top-decile seed valuations go to Bay Area or New York startups. Outside those markets, discount your valuation expectations by 20-30%.
- Time to close: 3 - 6 months for most founders. Plan your runway accordingly.
- Solo founder signal: Solo-founded startups now account for 35% of U.S. startups, but only 17% of VC-funded startups are solo-founded (Carta). VCs still strongly prefer teams.
- Option pool: Budget for a 10% option pool at seed. It is standard, and not modeling it creates nasty surprises.
What Kills Deals at the Last Minute
I see it constantly - founders losing deals not in the pitch, but in due diligence or in the weeks between a soft yes and a signed term sheet.
Four deal-killers that show up most often:
Cap table surprises. Stacked SAFEs at different caps that nobody modeled. Equity given to early advisors who didn't earn it. A co-founder who left with unvested shares. Clean your cap table before you start conversations. Investors will find everything.
Valuation anchoring to the wrong era. Founders who want to raise at 2021 multiples in the current market are not getting deals done. PitchBook analysis shows companies that raised at top-decile valuations without matching traction faced significantly longer Series A timelines - or couldn't raise at all. Benchmark your valuation against the past 6 - 12 months, not peak market conditions.
Desperation signals. Investors can tell when a founder has 60 days of cash left. The terms they offer reflect that calculation. Never start a process without 6+ months of runway. Founders who do get worse terms.
Overcomplication. Offering both a valuation cap and a discount on your SAFE gives investors double protection and makes your economics worse than a priced round. 83% of SAFEs use cap-only, per Carta. Keep it simple.
Building Your Investor List
Investors who will actually move your seed round forward are the ones who:
- Have written checks in your sector in the past 12 months
- Run a fund size that matches your round (1% - 2% of fund size = check size)
- Specialize at seed rather than occasionally participating
- Have portfolio companies that complement, not compete with, yours
A targeted list of 50 - 80 investors who genuinely fit those criteria will outperform a spray-and-pray list of 500 names pulled from a generic VC database every single time.
When you are building that list, you need to verify that each investor is actively deploying. A fund that made its last investment 18 months ago is not a real target regardless of how much their website says they love early-stage startups. Filter hard on recency.
Building that kind of targeted, verified contact list is exactly the kind of research that used to take a week and now takes an afternoon. Tools like ScraperCity let you search millions of contacts by title, industry, and company - so you can build a verified list of active seed investors without spending your limited runway on manual research. Try ScraperCity free with $5 in trial credit and see how fast the right list comes together.
After the Check Clears
Closing your seed round is not the finish line. It is the starting gun.
The moment money hits your account, the clock starts on your Series A. You have 18 - 24 months to generate the metrics that make a Series A a clear yes. Specifically: strong ARR growth, improving unit economics, and traction that demonstrates you found product-market fit.
I see it consistently - the founders who raise Series A rounds are not the ones who spent 12 months celebrating the seed close. They treated the seed wire as a 24-month countdown. Everything else followed from that.
Your investors are now your board or at minimum your stakeholders. Keep them updated monthly with honest numbers - good and bad. Investors who are surprised by bad news become difficult. Investors who are briefed on bad news with a clear plan to fix it become helpful.
Raise your next round when you have momentum, not when you are running low on cash. The best time to raise Series A is when you don't need it yet - when your metrics are trending up and you have 12+ months of runway remaining. That is when you get good terms.
The Short Version for Anyone Who Is Skimming
If you read nothing else, here is what is working right now:
Cold email at pre-seed works if you have traction and real personalization. Warm intros are faster. Send your champions a forwardable blurb they can copy-paste.
AI-generated decks are hurting founders. Show original thinking. Show the moat. Put the defensibility slide front and center.
Model your dilution before you start. Run the numbers on every SAFE. Use post-money SAFEs. Keep total seed dilution under 20%.
Target 18 - 24 months of runway. Start conversations 6 - 9 months early. Run parallel processes, not sequential ones. Get to no fast.
The lean team is now a signal. Six sharp people with clear roles beats twelve generalists with a bloated burn rate.
And persistence is the primary skill. The founders who close rounds are the ones who kept going after 100 rejections, not the ones who had the best deck.