Most important take away

Early-stage venture is not a trade but a long-term contract for alignment — founders should pick investors like co-founders, and investors should treat each check as one of a handful of multi-year commitments. Cem’s UiPath story (16.5M into ~2.1B realized) shows that valuation discipline matters less than founder quality, market size, and the willingness to bridge-fund overlooked companies when traction is heading the right direction.

Summary

Actionable insights for founders and investors:

• Treat early rounds as alignment contracts, not trades. Pick investors like co-founders; the cap table you assemble determines who is “rowing in your direction” for 5-10 years. The best founders rarely fill cap tables with only large multi-stage firms.
• Founder quality trumps everything. Cem ranks founder > market > traction. He has passed on great markets due to founder chemistry/ethics concerns and accepted founders who didn’t fit the mold (Daniel Dines wasn’t a polished English-speaking enterprise sales leader, but his earnestness and technical iteration speed won out).
• Price discipline matters but rarely changes power-law outcomes. Even if Cem had paid 2x at UiPath’s seed, it would still be a fund-defining outcome. Don’t lose deals on price for great founders, but don’t get rushed — categorically pass when the process is too compressed to understand the business.
• Use reserves for the misunderstood, not the obvious. Cem’s biggest insight from UiPath: when 40 funds passed and traction was quietly improving, internal conviction plus a bridge round (at a regretted $20M-ish convertible cap) became the highest-leverage check. By the billion-dollar B round, you are no longer in “overlooked” territory — you are riding momentum.
• Lean out strategically. Selling 1-10% of the position at each subsequent round (each at a premium to the primary) returned multiples of the fund before IPO. Post-IPO, early-stage investors lose their information edge — distribute in-kind and let LPs decide.
• Loss ratios that are too low can signal under-risking. In fund one, only 3 of 15 returned no capital — a “criticism” because some capital was deployed into safer mediocre returners instead of higher-risk, higher-return shots. Downside protection is overrated; liquidation preferences usually get renegotiated.
• Be quiet on the board. The asset chooses the investor in VC, which creates pressure to over-demonstrate value-add. Cem advocates obsessive board prep but carefully chosen interventions; sensitive topics should never debut in a board meeting — raise them 1:1 first.
• Watch for the premature scale trap. The fastest way to break a company is assuming product-market fit and scaling headcount and go-to-market scripts before iteration is locked in. The industry “fetishizes growth”; headcount is the easiest metric to inflate.
• Heuristic on valuation for founders: raise at a price you are highly confident you can 2-3x at the next round. Every 20% of dilution should buy you a >20% increase in probability of reaching the vision.
• Signaling is real. Examine cap table behavior — follow-on participation (or lack of it) reveals what insiders actually think. Founders should ask investors directly about their follow-on data and the cases where they declined to follow on.
• Concentrated portfolios beat spray-and-pray. 4-5 checks per partner per fund (15-18 companies total) forces care and ownership discipline (typical 10-20% first-check ownership).
• Vintage dominates fund returns. 2021 funds will mostly not return 1x. In murky transition periods (like the AI moment), the answer is not to sit out but to play your defined game with clarity about your edge.
• Career path note for aspiring investors: Cem moved from founder to angel to VC because the gap in his region’s funding stack (above $5-10M) put his angel investments at risk. Identifying a structural funding gap is a legitimate reason to start a firm.

Chapter Summaries

• Intro & path to venture: Cem started his first company in 1999 in NYC (social networking), exited, moved to Istanbul, became an angel in Turkish consumer internet, hit ~10x angel track with eBay and Delivery Hero exits. Started a fund out of fear that follow-on capital wouldn’t exist for his angel investments.
• On commoditization of VC: Disagrees with Doug Leone — early-stage isn’t commoditizing because the product is care, time, and attention, not cash. Multi-stage seed strategies look commoditized adjacent to asset-management businesses, but core early-stage practitioners (Benchmark, USV) stay disciplined.
• Founder choice & competing with multi-stage funds: The best founders pick investors as co-founders, not for the highest price. Cem will compete on $5M/$25M seeds when chemistry and alignment are right.
• Price discipline and power law: Even paying 2x at UiPath seed wouldn’t have hurt the outcome much. But operating with no valuation discipline isn’t possible — they pass on price when rounds get over-heated, with mixed hindsight results.
• Biggest misses: Bolt (Taxify) at seed — couldn’t see global trajectory or accept ride-hailing take rates. Lesson: be more open-minded on margin structure for exceptional founders.
• Founder > market > traction: Founder issues kill more deals than market issues. Market must be big enough to return the fund. Traction is a distant third at seed/A.
• Care, board seats, ownership: 4-5 investments per partner per fund. 10-20% first-check ownership target. Board seats when warranted.
• Signaling and scaling failures: Signaling exists and is real. Premature scale (headcount as vanity metric, rigid GTM scripts) is the #1 cause of sideways drift post-Series A.
• Board behavior: VCs over-broadcast value-add due to founder-chooses-investor dynamics. Be obsessively prepared, sparingly opinionated, and never surface sensitive topics first in a board meeting.
• Wealth and risk appetite: An LP challenged Cem’s high 10% GP commit, fearing it would make the team too risk-averse. GP commits should be proportional to wealth and career stage.
• Regional LPs: Most NA institutional LPs treat Eastern Europe as too exotic; mistake of applying PE regional-macro framework to VC, whose outcomes are global.
• UiPath deep dive: Met Daniel Dines in 2014 (DeskOver, 12 people, Bucharest). First check $1M of a $1.6M round at ~$7M pre. Bridged with $2.5M when 40 funds passed. Excel led A at $80M; B at $1B (added $10M); sold down 1-10% per round at premiums; exited rest post-IPO. ~85% of fund one’s 20x came from UiPath.
• Sell-down strategy & post-IPO: Started divesting at the $7B Series C. Believed early-stage edge decays post-IPO; distribute in kind.
• Institutionalization: Wants the firm to outlast him. Family offices don’t attract the team mission needed for institutional success.
• Losses and downside protection: Lowest loss ratio (3/15 in fund 1) can mean under-risking. Downside protection rarely changes outcomes — preferences get renegotiated.
• Liquidity markets & 2021 vintage: 10-year fund life is too short for seed/A. 2021 funds will mostly fail to return 1x. AI-era pricing concerns mirror 2021 frenzy — play your defined game.
• Quick fire: Early-stage check is a contract not a trade. Respects Fred Wilson, USV, Benchmark. Founders underestimate cap table leverage. Founders should ask investors about follow-on data — it reveals genuine care.