Euclid Insights: 2025 Year In Review
We often get asked by founders, LPs, and fellow GPs why we invest so much time and effort into content. Of course, there are material direct benefits to getting our name out there—we’ve discussed our belief that brand is more important than ever in the early-stage venture world, supporting sourcing and winning. But we’ve come to realize that the most powerful benefits of writing are harder to measure.
If you can’t explain it simply, you don’t understand it well enough.
— Albert Einstein1
Writing clarifies. As a two-person partnership, it could be all too easy to fall into shared, complacent patterns of thinking. Writing has been immeasurably important in sharpening our thinking, testing our reasoning, refining our frameworks, and iterating on our approach to early-stage investing—especially in a category as nascent and dynamic as Vertical AI.
Writing doesn’t just communicate ideas; it generates them.
— Paul Graham2
Writing forces continuous learning. In a field like private investing—where there are effectively no bounds on what you could know—the problem is often deciding what one should take the time to know. While research could be a never-ending task, writing requires commitment. Every essay is a time-bounded opportunity to dig into a worthy concept, uncover unknown unknowns, and guide our future learning.
The difficulty lies, not in the new ideas, but in escaping from the old ones.
— John Maynard Keynes3
Writing strengthens mental flexibility. Few people wake up every morning with the thought, “what am I wrong about today?” True intellectual honesty requires not only an openness to, but also an embrace of the possibility that you are wrong. Left unchecked, the mind will resolve dissonance long before it resolves truth. Writing is excellent practice in developing strong opinions, while holding them loosely.
How do I know what I think until I see what I say?
— E. M. Forster4
Writing makes us easier to know. Specialist and thesis-driven VCs often talk about the value of having a “prepared mind.” But what about founders? They’re expected to bear their soul as a matter of course, while investors remain intentionally obscure. We hope our writing introduces founders to our philosophy, approach, and expertise well before we meet.
Our belief is that these benefits extend well beyond the touchy-feely and into our ability to identify and support category-winners. And in a world where AI-generated content is easier than ever and the slopification of news feeds is accelerating… the value of clear, original, and engaging thought is most certainly on the rise.
This year, we set out to publish at least twice a month. So far, we’ve published 28 posts and recorded 8 podcast interviews. To recap the 2025, below are some highlights—both our most-read pieces and our favorites by topic.
As always, thanks for reading.
The Five Most-Viewed Essays
The five most-viewed posts on Euclid Insights, according to page views:
Dude, Where’s My Moat: Outlines why moats matter more in Vertical AI, evaluates prospective moats, and outlines a framework for planning and building defensibility from the earliest stages.
Market Sizing in Vertical AI (Part I): Challenges traditional TAM assessments, emphasizing the importance of understanding market dynamics and expansion potential in vertical AI. Part II helps founders to evaluate TAM and expansion strategies, considering ACVs and customer counts.
Emerging Playbooks in Vertical AI: Explores how vertical AI startups are leveraging domain-specific data and workflows to build defensible positions, transitioning from authoring layers to systems of intelligence.
We Have Met the Enemy and He Is Us: Reflections nearly a year after our initial piece on the damaging impacts of increasing venture capital consolidation on both innovation and the emerging manager ecosystem.
The AI-First Roll-Up: Analyzes the challenges VC-backed consolidators face in roll-up strategies and introduces AI-driven synthetic software roll-ups as a more viable alternative.5
Additional Favorite Essays
We also wanted to highlight some of our other favorites from last year by category:
Founder Playbooks & Toolkits
Strategies in vertical company building.
Voice-First Playbooks in Vertical AI: Presents a thesis on the potential of voice AI to redefine software interfaces, emphasizing the importance of reimagining workflows with voice as the primary modality.
A Guide to Disrupting Incumbents: Outlines strategies for startups to weaken incumbent positions by reducing complement prices, switching costs, or direct demand.
The Truth About Founder-Led Sales: An exploration of why founder-led sales are crucial in vertical software startups, detailing the stages of product-market fit and the risks of premature delegation.
Market Data & Dynamics
Data-driven analysis of trends in Vertical AI.
Deus Ex CapEx: Examining the assumptions and incentives behind the economy-wide bet on GPUs. Are we in a bubble? Are app-layer founders seeing the benefits of this spending spree in AI pricing?
Vertical AI & the Productivity Paradox: Discussion of the “AI will kill jobs” narrative plus an analysis of where and how Vertical AI can impact labor productivity & produce significant venture outcomes
Does AI Threaten Vertical SaaS?: Explores how AI-induced commoditization affects vertical SaaS, identifying strategies for companies to maintain competitive advantages.
Investment Perspectives
The overlap of VC, capital markets & vertical enterprise.
The Quiet Death of VC-Founder Alignment: A discussion of how AUM and AI have shifted investor incentives, and why founders should pay attention. Why your $1B exit may not actually move the needle.
Swiss Cheese & Vertical Software: Introduces the WYHTB framework, applying the Swiss Cheese Model to assess multivariate risks in Vertical AI startups, emphasizing the importance of aligning organizational and market factors to achieve success.
Early-Stage VC in the Age of Vertical AI: Discusses the necessity for early-stage investors to adopt first-principles thinking in evaluating Vertical AI opportunities, focusing on long-term TAM.
Our Show: VERTICALS
This Fall, we co-launched VERTICALS—a weekly video show & podcast—to give Vertical AI the space it deserves. Co‑hosted by Luke Sophinos and Euclid’s own Nic Poulos, each episode covers a digs into strategy with a top founder and covers a market pulse including latest financings, exits, and thought pieces. Join us here to get new releases in your inbox every Wednesday:
Here’s a wrap up our episodes so far:
#9 — Vertical Business-in-a-Box
Dan Friedman (Founder) @ Moxie, Meadow Memorials, Boulton & Watt
Drawing on his success with Moxie, Dan breaks down how to build repeatable, operationally tight vertical “business-in-a-box” startups.#8 — Vertical Aggregators
Sam Youssef (Co-Founder, CEO) @ Valsoft Corporation
Sam explains Valsoft’s disciplined approach to acquiring, operating, and compounding 100+ vertical SaaS businesses.#7 — Vertical SMB Go-to-Market
Kyle Norton (CRO) @ Owner.com
Kyle dives deep into what actually works when selling to vertical SMBs—why distribution, trust, speed, and the right team matter most.#6 — Healthcare’s AI Moment
Mike Kopko (Founder, CEO) @ Pearl Health
Mike outlines how AI is finally reshaping healthcare delivery by aligning incentives across payers, providers, and patients.#5 — Vertical Community Building
Todd Saunders (Founder, CEO) @ Broadlume
Todd shares how Broadlume turned community, content, and data into a durable moat within the fragmented flooring industry.#4 — AI Marketplaces
Jack Greco (Co-Founder) @ ACV Auctions
Jack walks through the evolution of ACV from a simple marketplace into a data- and AI-driven platform reshaping the wholesale auto industry.#3 — Hyper-SMB Strategies
Brian Litvack (Co-Founder, CEO) @ LeagueApps
Brian discusses how LeagueApps built a category-defining platform by obsessing over the long tail of SMB operators running youth and local sports organizations.#2 — AI Roll-Ups
Jeremy Yamaguchi (Founder, CEO) @ Cabana
Jeremy unpacks the AI-first roll-up thesis—why software, automation, and vertical focus change the economics of acquiring / operating service businesses.#1 — Unbundling the Franchise
Ilir Sela (Founder, CEO) @ Slice
Ilir explains how Slice is unbundling the traditional franchise model by giving independent pizzerias enterprise-grade tools without taking away ownership.
Join the debtate, wherever you watch / listen: 📺 Youtube | 🎧 Spotify | 📖 Substack
Our Biggest Influences This Year
Bessemer Venture Partners (Team)
BVP’s work on vertical software reinforced a core belief: enduring outcomes come from trading market size for market share—and compounding that advantage across workflow and data.Emergence (Team)
This firm has been an inspiration for Euclid in so many ways. Since their “industry cloud” thesis, they’ve been pioneers of vertical thinking. Jake, Kevin, Joe, Yaz—nearly the whole team have been career-long guides, friends, and influences. We’ll add, Jake (with his thoughtful, sober AI takes) and Joe (with his less-sober but equally smart takes) are must-follow voices on LinkedIn.Sequoia Capital (David Cahn)
Their writing is infrequent but invaluable (recall “RIP Good Times”). This year, Cahn’s analysis of AI infrastructure economics shaped our take on the CapEx-heavy AI investment boom and or related piece (see above).Greylock (Mike Duboe)
The firm has been a strong influence for Euclid. This year, Duboe’s writing and operator-centric perspective sharpened our view on AI-native company building—especially re: where workflow leverage creates defensible advantage.Paul Graham
Graham is the second godfather of venture/startup writing. He took what Fred Wilson started and turned it into an art. This year, revisiting his work reinforced a foundational principle for us: writing as a tool for knowledge and improvement.Altimeter (Jamin Ball)
Clouded Judgement is, of course, a must-read. Ball’s clear, data-driven breakdowns of AI economics and weekly public-market signals help anchor our thinking and consider venture-land in a larger capital markets context.
Favorite Substacks
Linear (Luke Sophinos)
The godfather of vertical software writing—consistently covering the vertical ideas that matter (and, not coincidentally, our illustrious VERTICALS co-host).Ali Afridi (Sandhill.io)
A meta-choice: consistently aggregating some of the best VC and founder writing in our orbit, with a bias toward signal over noise.Epoch AI (Team)
Essential reading for staying close to the true frontier of AI capability—cutting through hype with rigorous, research-driven insight.Matt Brown (Matt’s Notes)
Exceptional vertical fintech thinking paired with a rare gift for visualization—turning complex system dynamics into ideas you can actually reason about.Eli Dukes (Verticalized)
Another vanguard of all things vertical. Deep lens on how real industries actually change with an outlook we respect: if you’re doing market maps, you’re too late.Colin Gardiner (Take Rate)
One of the sharpest lenses on marketplaces and transactional businesses—an area we think is quietly being reshaped by AI, and one we’re watching closely.Jackie DiMonte (Day by Day)
50% of the crack team at Grid Capital, Jackie is fellow new entrant to Substack that you may know from her incredible data & charts. Tufte would be proud!
Favorite Commentators
Gokul Rajaram (Marathon)
Everyone knows and loves Gokul for a reason. He’s connective tissue in our world—and as he’s posted more this year, it’s a gift: never long, always worth reading. We hope to see more from him and his equally talented partner Mike Gilroy (formerly of Coatue) at Marathon in 2026.Alex Niehenke (Scale Venture Partners)
A great commentator on Vertical AI from a firm that’s quietly carved out real credibility in the space. Alex is a longtime friend and thought partner—and it’s been great to see him sharing more of his thinking publicly.Todd Saunders (Founder, Broadlume)
The grand vizier of flooring and a true champion of vertical SaaS founders. Ever-present on vertical LinkedIn for good reason—podcast guest, portfolio helper, and one of the most constructive voices in the space.Rick Zullo (Equal Ventures)
A fellow vertical obsessive and old friend, Rick never holds back—their thesis-driven, prepared-mind mentality is one shared here at Euclid. It’s all too rare to see firms openly sharing deep research as they do.Charles Hudson (Precursor Ventures)
Quieter on LinkedIn, but consistently thoughtful in his Substack notes. A steadfast champion of the inception and pre-seed stage. No AI slop, just hard-earned, honest takes. Also has a worthwhile podcast in The Pitch.
Attributed in The New Quotable Einstein (2005)
The Need to Read (2022)
Aspects of the Novel (1927)
Initially published in late 2024, but remained one of our most viewed essays in 2025.


