Why Margins Don't Matter for Early-Stage Startups | Gili Raanan

Why does it matter? Because most venture math is a losing bet — and almost no one admits it.

Gili Raanan built one fund with a decacorn, seven unicorns, and three acquisitions from 19 companies. His read on the current venture market isn't cautiously pessimistic — it's structurally damning. The entry prices have ballooned while the actual output of winners hasn't budged, and the people writing checks broadly are the ones who'll get hurt most.

• Only 1-2 cybersecurity unicorns emerge per year — 2021 was the exception, not the new baseline
• LPs distributing venture allocations broadly are running negative-expectation bets without knowing it
• Gross margins are irrelevant for early-stage companies — the right conversation is in 2029
• Fast growth becomes company DNA and almost never fades without a major external shock

The venture market isn't broken — it was never supposed to work for most people in it

Entry prices at the seed stage have ballooned while unicorn output holds flat at one or two per year. Do the math on 150 new cybersecurity startups funded annually in Israel alone, multiply by 2.5 to get the global number, and the hit rate is still 1%. "The likelihood you'll hit a successful company is still one to 1%. It's one out of 150, maybe two out of 150. And the prices the entry prices where you buy stock at the seed stage is going significantly higher."

Raanan doesn't hedge this. "I think it's going to end up with some serious catastrophe for many of the players." If you're an LP who spread your venture allocation broadly, he says, "I wouldn't sleep well at night."

The counterargument — that bigger outcomes justify higher entry prices — he accepts intellectually and dismisses practically. Crowdstrike and Palo Alto reaching never-before-seen market caps doesn't change the probability math. It might shift the ceiling on winners, but the floor on losers stays the same. The only rational response: founders should pick financing partners with more care, and LPs should concentrate rather than diversify across the venture landscape. The game rewards concentration. Spreading bets looks like prudence and functions like slow-motion value destruction.

Gross margins are a 2029 problem — obsessing over them now is how you kill a company early

Raanan never discusses gross margins with his early-stage portfolio companies. Not occasionally. Never. "How often I discuss gross margins with my early stage companies. Never."

His logic is sharp: every stage of a company's life demands different problems be solved, and forcing founders to optimize for margin before they've compounded growth is like asking someone to worry about retirement savings before they've landed a job. "Gross margins are important. Let's talk about it in 2029 and let's build the foundations of a healthy business — assuming that we would get to deal with gross margins."

That last phrase carries the real weight. He's not saying margins don't matter. He's saying the assumption that you'll get to the margin conversation is itself a stretch goal. Most companies never reach the scale where margins become the defining variable. The ones that do earned that problem by growing ferociously first.

On AI specifically, he's honest about the limits of his knowledge — there isn't enough track record with profitable AI businesses to draw reliable vital signs. But his instinct is that gross margins will remain important there too, eventually. The variable isn't whether margins matter — it's when.

Fast growth isn't a metric — it's a character trait that compounds

Sierra, one of Raanan's portfolio companies, sold half a million in its first quarter, a million in the second — then literally zero for two quarters. Most investors would have written that off. Instead, the founding team diagnosed the stall, adjusted, and went on to book $12 million of new business in the next twelve months.

That pattern — explosive start, near-death plateau, re-ignition — didn't shake Raanan's conviction because of something he learned across dozens of investments: "Whenever a business is getting to a point it's growing super fast year over year, it becomes part of their DNA. So it would not slow down just because of averages. There needs to be a significant external event to slow them down."

This is a sharper claim than it sounds. It reframes early growth signals from lagging indicators of product-market fit into forward-looking signals of organizational character. "That thing that makes them move so fast — typically, in most cases — would not simply fade away." The product will change. The market positioning will shift. But the operational instinct to grow fast is wired in early and persists.

Island — now a $5 billion company selling enterprise browsers — is his clearest example of the other side: a market that didn't exist in 2019, zero customer demand signals, and explosive growth anyway. No prior framework would have caught it.

An IPO is a marketing event — and treating it as an exit is a category error

"Going public is not a financial event. It's a branding event." Raanan is blunt: an IPO is the moment a company tells customers, partners, employees, and future recruits that it intends to stick around. That's the entire value of it.

Everything else about the IPO is a cost. Lock-up periods, disclosure requirements, analyst scrutiny, trading restrictions — "it's hell for liquidity, but it's an important marketing event." The IPO doesn't unlock capital or provide founder exits. It signals permanence to a market that needs to believe in you before it buys from you at scale.

This reframes the entire staying-private-longer debate. The question isn't whether founders are avoiding the public markets out of reluctance or immaturity. It's that the public markets solve a trust problem, not a liquidity problem. And if liquidity is what you actually need — for employees, for early investors, for LP returns — the solution is structured secondaries, not an IPO timeline.

Fully vested employees are structurally forced to leave — unless you build a recurring secondary program

Late-stage private companies have a hidden talent retention crisis baked into their structure. Four-year vesting schedules make sense at inception. By year six or seven, your best engineers and product managers are fully vested, hold most of their net worth in one illiquid stock, and face a rational decision: diversify by joining a new company with fresh equity, or stay locked in a paper position with no liquidity mechanism.

"You actually force them out of the company," Raanan says. The exit isn't driven by dissatisfaction — it's driven by personal finance logic.

Cyberstarts built a direct solution: an employee liquidity fund that underwrites a recurring annual tender offer across portfolio companies. The goal isn't a one-off secondary transaction. It's creating the expectation of regular liquidity — the same relationship employees have with public market shares — inside a private company structure. "The employee of that company knows that they are getting liquidity. The very same type of liquidity they would get in a public market, they would get in a private company."

The first program launched with Sierra, covering hundreds of employees for many millions of dollars. The model: management sets the price, Cyberstarts underwrites, other investors can participate. No hoarding of deal flow, no adversarial cap table dynamics. Just a structural fix to a structural problem.

Venture is the science of exceptions — pattern-matching from past winners is how you miss the next one

No-Name Security hit $3 million in year one, $15 million in year two, then stalled. The API security market turned out to be a niche inside application security, not a standalone category. Akamai bought it for around $500 million. Good outcome. Wrong trajectory.

Island, founded the same year, built a product no CISO said they needed — an enterprise browser competing against free tools from Google and Microsoft. Now it's a $5 billion company with Fortune 100 financial services clients replacing Chrome and Edge at the infrastructure level.

"The conclusion in my mind is that we are exercising the science of exceptions. It's good that we share those lessons, but if you just take those lessons and apply them linearly, I think it would be very hard for you." His advice to younger investors feeling exposed by the collapse of prior frameworks — rule of 40, triple-triple-double-double, early margin focus — is direct: absorb the lessons from people who've been through it, then trust your gut. "Nobody knows better than you do."

Rigid frameworks produce median outcomes. The exceptions — the ones that define careers and funds — almost always violate the rules that the prior exceptions created.

Fix weaknesses and you reach the market average — amplify strengths and you create real advantage

The instinct in building a venture partnership is to identify each partner's gaps relative to a known template and close them. Raanan thinks this is exactly wrong.

"I would let each team member play on their relative strengths and would not require them to focus on improving their weaknesses." The ceiling on weakness improvement is market parity. "On their weaknesses, at the best case they can be as good as the market. But on the things that are exceptional, they are creating real advantage, real beta."

The practical result at Cyberstarts: a team where every partner operates differently, and that difference is respected rather than corrected toward a single standard. Chemistry and mutual respect hold it together. Uniformity doesn't.

The real liquidity infrastructure for venture hasn't been built yet

The extension of private markets isn't a stopgap — it's a permanent structural shift. IPOs as branding events, recurring tender offers as the actual liquidity mechanism, secondary programs built into operating models from early growth stages: these aren't workarounds for a broken system. They're the system replacing the old one.

What this episode quietly reveals is that the firms building proprietary liquidity infrastructure — not just investing, but controlling when and how employees and early investors get liquid — are accumulating a compounding retention advantage their competitors can't easily replicate. The fund that underwrites the annual tender offer owns the relationship in a way that a passive cap table participant never will.

---

Topics: venture capital, cybersecurity, early-stage investing, fund economics, secondaries, talent retention, IPO, product-market fit, portfolio management, Israel tech