Drew Cohen on Adobe's AI Threat, Constellation Software, Copart's Moat and Shift4
TL;DR
Drew Cohen of Speedwell Research argues that Constellation Software's 50% drawdown due to AI fears is overblown, explaining that mission-critical vertical market software customers prioritize reliability and switching costs over marginal cost savings, making disruption by AI-generated alternatives unlikely despite the technology lowering barriers to entry.
📉 Constellation Software Overview 3 insights
Historic valuation opportunity
The stock has fallen over 50% to a $39 billion market cap due to AI disruption fears, creating the largest drawdown in company history despite consistent mid-teens revenue and free cash flow growth.
Vertical market software conglomerate
The company owns 70,000+ niche software businesses serving mission-critical functions like cemetery management and bus scheduling, typically with small TAMs and high customer concentration.
Strong underlying economics
Current valuation sits at 21x free cash flow with high-teens to low-20s return on invested capital, supported by $2+ billion in annual free cash flow generation.
🤖 The Flawed AI Threat Thesis 3 insights
AI doesn't change competitive dynamics
Most vertical software was already simple to build (1-2 engineers, 6 months), so AI accelerating development to days or weeks doesn't materially alter the competitive landscape for incumbents.
The 10x improvement rule
Citing Peter Thiel, Cohen argues disruption requires a 10x better product, not marginal improvements, because switching costs for mission-critical systems are prohibitively high for mere 20-50% cost savings.
Distribution remains the moat
Serving 70,000 micro-markets requires extensive salesforces and relationship-building; AI cannot replicate the distribution network needed to reach niche customers like chicken coop operators or municipal bus schedulers.
đź”’ Why Customers Won't Switch 4 insights
Reliability trumps cost savings
Business owners' highest preference is that software never breaks, as downtime causes permanent revenue loss, making them unwilling to risk unproven AI alternatives regardless of price.
DIY software carries catastrophic risk
While business owners could use AI to build custom solutions, the asymmetry of risk—potential total business loss from uncorrectable code errors versus modest software costs—makes self-built options economically irrational.
Embedded professional services
Constellation generates significant revenue from customization and support, embedding engineers into client workflows and creating compliance-grade gold standards that are difficult to displace.
Legacy infrastructure inertia
Many customers still operate on-premise servers and have only recently transitioned to SaaS, indicating low technology risk tolerance and preference for outsourced IT responsibility rather than DIY approaches.
Bottom Line
Constellation Software's 50% drawdown presents a compelling entry point because AI tools fail to overcome the fundamental switching costs, reliability requirements, and distribution challenges inherent in mission-critical vertical market software.
More from We Study Billionaires (TIP)
View all
OTC Markets (OTCM): A Picks and Shovels Play in Modern Capital Markets
OTC Markets Group operates as a quasi-monopoly infrastructure provider for over 12,000 over-the-counter securities, serving as the essential 'picks and shovels' play for small-cap capital markets. The company has generated exceptional shareholder returns through a capital-light model that has compounded free cash flow at 14% annually for a decade while maintaining zero debt and operating margins exceeding Alphabet.
Why This Real Estate Data Empire is Making a $5 Billion Bet
CoStar Group, the dominant commercial real estate data company with an unassailable 37-year data moat and 59 consecutive quarters of double-digit growth, has invested $5 billion to enter the residential market with Homes.com, sparking a 50% stock selloff and activist investor revolt over concerns of capital misallocation.
Charlie Munger's Secret to Beating the Market w/ Ryan Sablan
Ryan Sablan shares his asymmetric value investing framework, revealing how Charlie Munger's simple mathematical insight—that you only need one winner among three bets to achieve 33% returns—combined with rigorous balance sheet analysis and position sizing, can lead to market-beating performance without requiring a high win rate.
Portfolio Review: Thoughts on Airbnb, Reddit, Adobe, and co.
The hosts conduct a quarterly review of their concentrated 17-stock portfolio, announcing plans to sell two positions and add to one, while deep-diving into Exor's 60% discount to NAV as a Ferrari proxy and debating the future of their small TransDigm stake.