Oct 27, 2023 · 50m · news
Ed Sim: Why Seed Has Never Been More Competitive & Why Pricing Has Never Been Higher | E1076 · 20VC with Harry Stebbings
0:00 / 0:00
▶ Watch on YouTube →
gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions
In this episode of 20VC, host Harry Stebbings interviews Ed Sim, founder of Boldstart, about the mechanics of "inception investing" and the hyper-competitive seed-stage landscape. Sim critiques the dangers of over-capitalization, explains how to navigate the AI hype cycle, and shares key lessons on portfolio conviction using real-world case studies like Snyk.
How this conversation actually went
Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. Harry holds 23% of the talking time here. How this is scored →
Harry as informed peer 5.5 Guest teaching 5.3 Guest disagreement 2.1 Harry pushing back 4.8
speaking balance: gold is Harry, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 19:34 Ed counters SaaS uninvestability claim
Ed directly rejects Jason Lemkin thesis brought by Harry that enterprise SaaS is uninvestable, pointing forcefully to re-accelerating IT spending and cloud migration numbers.
Hardest push from Harry ▶ 4:23 Harry forcefully rejects mega seed roundsHarry strongly rejects the market practice of 20 on 100 seed rounds, calling it a harmful practice and citing his debate with Parker Conrad.
Biggest teaching moment ▶ 47:11 Ed breaks down interest rate valuation compressionEd educates the host using JP Morgan data showing how interest rate hikes crushed forward software revenue multiples from 35x down to 6.8x.
Harry holds his own ▶ 39:15 Harry critiques pro-rata reserve strategies with counter-examplesHarry demonstrates high domain expertise by challenging reserve allocation strategies, citing Hopin, Clubhouse, and BeReal as examples of misleading initial traction.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | Harry as informed peer | Guest teaching | Guest disagreement | Harry pushing back | Why |
|---|---|---|---|---|---|---|
| Introducing Ed Sim and Inception Investing | 3 | 6 | 2 | 3 | Harry challenges whether inception investing is simply a new label for pre-seed. Ed educates him with PitchBook data showing pre-seed median age is 1.2 years versus 2.7 years for seed. | |
| Why Seed Valuations and Round Sizes are Inflating | 7 | 2 | 2 | 8 | Harry aggressively rejects giant early seed rounds, calling them harmful and citing his debate with Parker Conrad on capital efficiency. | |
| Navigating Multi-Stage Competition and Valuation Pressure | 5 | 6 | 3 | 5 | Harry questions if founders genuinely desire round size constraints. Ed breaks down the three inception round structures: discovery, classic, and jumbo. | |
| The Dangers of Jumbo Seed Rounds and the Superhuman Lesson | 5 | 5 | 1 | 4 | Harry and Ed discuss how excessive runway damages founder discipline, with Ed illustrating Rahul Vohra strategy at Superhuman. | |
| Sizing the Modern Seed Fund and Snyk's Multi-Round Strategy | 7 | 6 | 2 | 7 | Harry uses math to argue traditional seed fund sizing is broken, forcing Ed to reframe optimal fund sizes between $150M and $250M using Snyk as an example. | |
| Board Cadence and Snyk's "Jesus Moment" | 7 | 7 | 4 | 7 | Harry cites Jason Lemkin argument that enterprise SaaS is uninvestable, but Ed counters with data on re-accelerating IT spend and cloud migration stats. | |
| Dilution Realities and the Opportunity Fund Strategy | 6 | 4 | 3 | 6 | Harry brings up his viral tweet criticizing pre-seed board seats, prompting Ed to explain his flexible founder-led quarterly cadence. | |
| Navigating the AI Hype Cycle and Bubble Parallels | 5 | 5 | 2 | 4 | Harry and Ed analyze whether seed AI deals represent a bubble, comparing domain hype to the dot-com era while noting 80% of portfolio companies deploy AI features. | |
| Snyk's AI Strategy and the Outlook for Tech M&A | 6 | 6 | 3 | 6 | Harry posits tech M&A is frozen due to regulatory and scale issues, but Ed counters that growth funds will take 1x returns, using Loom as evidence. | |
| Valuations in the Growth Market and Advising Founders | 5 | 5 | 1 | 3 | Harry asks if growth markets are dead, and Ed cites Carta data on Series C valuation drops while advising founders to prioritize fair pricing. | |
| Sizing Inception Rounds & Multi-Stage Option Checks | 5 | 4 | 2 | 4 | Harry brings up inflated valuation decks from pedigree firms, and Ed explains how multi-stage funds view multi-million dollar checks as simple options. | |
| Venture Capital Lessons on Excess Cash and Preemptive Rounds | 8 | 6 | 3 | 7 | Harry challenges automated pro-rata reserves by pointing to deceptive short-term traction in Hopin, Clubhouse, and BeReal versus compounders like Figma and Snyk. | |
| Ed Sim's Greatest Hits: Backing Customer and Snyk | 4 | 7 | 1 | 2 | Ed details his investments in Customer and Snyk, detailing the operational realities of competing with incumbents and category creation. | |
| Quick-Fire: The Real Meaning of AI and Venture Irrationality | 6 | 7 | 2 | 5 | In a quick-fire sequence, Ed shares JP Morgan research mapping interest rate hikes directly to software forward multiple compression from 35x to 6.8x. | |
| Boldstart's Future and Pure Inception Focus | 3 | 3 | 1 | 1 | Ed outlines Boldstart long-term firm architecture, explicitly rejecting asset management expansion to remain focused on inception rounds. |
Statements from this episode (45)
Sim: Inception rounds split into discovery, classic, and Megatron jumbo tiers
“The point is the three rounds are, there's a discovery round, which my opinion is less than two million. So it's usually like maybe a first time founder exploring a new market. It would graduate to the next round, which would be a classic. And a classic round,…”
Sim prefers structuring Boldstart inception rounds at $3M to $4M
“I would prefer Your classic structure of three to four million dollars, to be honest with you.”
Sim: Median pre-seed company age is 1.2 years, seed is 2.7
“The median age of a company that raises a pre-seed round is 1.2 years of age now. And the median age of a company that raises a seed round is 2.7.”
Sim: Repeat founders avoid pre-seed rounds to prevent extra equity dilution
“A lot of times founders who are second and third time founders don't want to even be called to have a pre-seed round because it presupposes you need a seed round. And what does that mean to you, Harry? It means that if you have a pre-seed round and you have a …”
Stebbings: Ex-OpenAI and Stripe staff raise $20M on $100M valuations
“You have person leave stripe or leave open AI or hugging face, and they raise 20 on a hundred. And that's happening more and more.”
Ed Sim: Billion-dollar multi-stage VC funds stopped growth investing in 2022
“The spigot dried out last year, all the big multi-stage firms with a billion dollar plus funds stopped investing in growth.”
Ed Sim: Global quarterly VC funding dropped from $200B to $73B
“If you look at the peak, I think the peak was Q four, 20, 21, that about two hundred billion dollars was invested. Most recently it was seventy three billion in the last quarter.”
Stebbings: Too much early cash harms 99.9% of startups
“I think that too much cash too early Has a net negative impact on 99.9% of companies unless you are an exceptional allocator of capital like he is, like Daniel Ek is, like very, very few generational founders are.”
Sim: Boldstart pushes down seed round sizes to test founders
“I'll lead you around now if you want three, but not at six. So they come back to us, you know, a few days later and like, you know what? How about four? And the reason I do that Harry is because I want to test them. I want to test them to make sure that they a…”
Ed Sim: Boldstart will split $8M–$10M seed rounds for exceptional founders
“The way that we've evolved now is that sometimes I have to play ball, Harry, right? Sometimes there are some exceptional founders where, okay, you know what? I will split. An eight to ten million dollar round for that exceptional founder, depending on what tha…”
Ed Sim: Incubating companies is unscalable unless it is a firm's sole focus
“And by the way, this is not incubating a company. That's not a scalable model. And that's, unless that's all you do.”
Sim: Superhuman founder locked $10M Series A in separate bank account untouched
“Then he raised pre raised a ten million dollar A round. He put it in a separate bank account and said, I'm not going to touch this thing. I'm literally just going to put it there. I don't want to see it. It's in a separate bank account. I'm only going to opera…”
Sim: Boldstart Writes $4M-$6M Follow-On Checks for High-Executing Portfolio Companies
“When I start with a three and a half, four million dollar check, I'm writing a four to six million dollar check in the next round if the company's executed.”
Sim: Boldstart funded Snyk through three pre-Series A capital rounds
“You know, guy from sneak, Harry, do you know how many rounds he raised before? He raised his A three rounds of capital before his A. Guess who gave him all three rounds?”
Sim: Snyk Would Have Failed Without Continuous Internal Funding Prior to Series A
“If Guy went out trying to raise capital from someone else and I didn't have enough capital to give to him, he would have gone out of business.”
Sim: Optimal Modern Seed Venture Funds Must Be $150M to $250M
“I think a fund size between one 52 50 for doing this in a highly concentrated position with ownership and doing it in a way that allows you to write bigger checks after the discovery round to help a classic round, you know, raise a little more of an extension …”
Sim: Boldstart launched as a $1M fund in 2010
“We remember we started out with a million dollar fund. And bold start one. This was in 2010 when seed was the first round of record.”
Sim: Seed and pre-seed investors take less risk over time
“And then the pre-seed category was created because seed, seed investors weren't taking as much risk. And now the pre-seed investors aren't taking as much risk, which is why I'm saying let's do inception investing.”
Ed Sim: Enterprise IT spending is re-accelerating
“Enterprise IT spending is re-fucking accelerating right now.”
Ed Sim: Only 25% of cloud migration is complete
“We're still only 25% of the way of migration from on-prem to Cloud right now”
Ed Sim: JPMorgan Chase spends $1 billion annually on cybersecurity
“And JP Morgan, for example, spent a billion dollars a year in cybersecurity.”
Sim: Boldstart will never fund AI startups spending mostly on Nvidia
“You will never find both start, for example, finding companies where, you know, they're raising 20 or thirty million dollars, and two thirds of that money goes to NVIDIA.”
Ed Sim: Pre-seed startups raising $500K-$1M don't need a board
“If you're raising 500 K to a million, that's what I think of pre-seed, then yeah, you probably don't need a board. Yeah. Yeah. You don't need a fucking board.”
Ed Sim: Startups raising $3M-$4M should avoid SAFE notes for priced rounds
“If you're doing a classic round of you're raising three to four million, I like to tell the founder that, hey, you should probably not do a series of safe notes here because you never really know what you own, especially in conversion. You want to actually do …”
Ed Sim: AI is transformational but making money will be extremely hard
“I think this whole AI thing, frankly, I think AI is the most transformational thing that we've, that we're ever going to see in a long time. However, I still think it's really fucking hard to make money there.”
Ed Sim: Boldstart will not fund AI seed rounds at $100M post-money
“Chasing anything with a dot AI in their domain or an AI thing, I think is insane. However, I do think there's some opportunities there, but I'm not going to be funding these things at a hundred posts.”
Ed Sim: AI hype mirrors dot-com bubble and $100M seed portfolios will fail
“I just think that this is just like I saw in the internet bubble. Everything had a .com on it back in it, because I saw that. I lived through that shit. And we're gonna see the same thing. There are gonna be some massive companies built, no doubt, but it's not…”
Stebbings: 85% of 20VC deal flow is AI-first
“I now see 85% of our deal flow now purely as AI first deal flow.”
Sim: 55% of Boldstart portfolio companies feature AI offerings today
“Currently in the portfolio of our companies that are running from funds four through six, 55% of our portfolio companies have an AI related offering today.”
Sim: 80% of Boldstart portfolio will offer AI within 12 months
“Within the next 12 months, okay, 80% of the portfolio will have an AI offering.”
Harry Stebbings: Tech M&A market will stall across all deal sizes
“So I think M&A is going to go to shit. I think anything sub a billion is not meaningful enough for M&A teams and corp dev teams to actually engage with. And I think anything over a billion is going to be incredibly arduous to get over the line from a regulator…”
Ed Sim: Tech M&A volume will increase as investors seek liquidity
“I think we're going to see a lot more M and A's kind of coming ahead right now.”
Ed Sim: Growth funds returning 1x on peak deals will outperform peers
“So I guess what I'm trying to say is I think those growth kind of funds that put the money at the peak I think the ones that deliver one X or slightly above will probably do very well. We'll be above, you know, cause there'll be a lot of ones that are underwat…”
Ed Sim: Top Infrastructure Startups Still Raising Growth Rounds at $250M-$400M Valuations
“There's still pockets Of opportunity where people are going from one to three to four million in the infrastructure space that I see to 10 to 12 kind of on the forecast and they're still raising at two 50 to 400 now.”
Stebbings: VCs should strategically sell equity over time instead of doubling down
“My biggest lesson from the last few years was actually just to sell at some points, you know, we were always told, lean in the bullshit. You need to lean out strategically over increments over time.”
Stebbings: Allocating reserves based on early traction misses generational enterprise companies
“What's working often is unsustainable, and it's your sneaks, your figmas, your big IDs, which actually may take a bit longer. But you wouldn't allocate to them if you were doing it on traction.”
Sim: Startups solving hard technical problems appear slow before hitting rapid growth
“Some of the hardest problems to solve will take longer and it'll look like your slowest plotting company to start with. But in the long run, to your point, these things are the ones that can actually go, you know, hit the J curve pretty quickly.”
Startups taking on incumbents must build baseline table-stakes features first
“No matter how, how, how big of a vision you have, you still need at least the table stakes of, you know, check box of four things that they may need before they even believe in your vision. So it takes longer.”
Enterprise software markets repeat and rebuild identical tools every 10-15 years
“So that's what happens in enterprise. It's the same shit getting rebuilt every 10 or 15 years in a different way with a different pitch.”
Sim: Snyk took nearly three years to reach $1M in ARR
“It took him two and a half years to kind of get out there and get his first, you know, two and a half to three to get his first million of ARR.”
Ed Sim: Venture money chasing LLMs will be vaporized
“Let's just say the money chasing all these LLMs, there'll be a lot of money vaporized for sure.”
Ed Sim: $30M to $50M inception rounds are bonkers and irrational
“I mean, as I said, I'm looking at the inception rounds, right? I laid out a framework from zero to zero to six, zero to eight, zero to 10. But you know, you're getting 30, 40, fifty million dollar, you know, inception rounds done. And I think it's just bonkers…”
Ed Sim: Tech IPO window won't reopen until Fed cuts interest rates
“I think it's very simple. Mr. Jerome Powell signaling to the world that interest rates are under control now. And the air has been taken out of the balloon and the interest rates have to go down.”
Shorts cut from this episode
▶ Is Enterprise SaaS Investing Back? · 20VC with Harry Stebbin (@20:07)
▶ Why Talent AVOIDS High Valuation Startups 🙅♂ · 20VC with H (@5:20)
▶ Why the best founders DON’T want the highest valuation · 20V (@34:19)
▶ Are growth rounds dead? 💀 · 20VC with Harry Stebbings (@32:13)
▶ AI = Dot-Com Bubble? · 20VC with Harry Stebbings (@25:30)
▶ Why I invested $10M+ in this AI security company — Ed Sim · (@7:24)