When Accel invested in Cursor, its ARR was just $100K. They projected it would hit $300K by year-end; it hit billions. This experience shows that for generational companies, obsessing over financial projections is futile. The astronomical financials are merely a reflection of an unprecedented product-market fit that can't be captured in a spreadsheet.
When evaluating board members, founders should be wary of those who are the most vocal. There is an inverse correlation between how much someone talks and how helpful they are. The board members who feel a need to always have the first and last word are often less insightful than those who listen and offer concise, thoughtful observations.
Accel partner Miles Clements assesses AI companies by how quickly users get value and how long that value lasts. He notes that the best vertical AI, like coding assistants, excels on both dimensions: offering immediate productivity gains that compound over time, making the value highly durable.
Despite public claims from figures like Chamath Palihapitiya that 'Cursor is dead,' Accel's Miles Clements reveals the company's agent product grew 15x with 90% DAU. This highlights the disconnect between social media sentiment and actual product traction, proving that internal metrics are the ultimate source of truth.
Instead of being a weakness, Cursor's reliance on multiple foundation models is a key strength. With 50% of developers switching model families daily, this approach allows Cursor to benefit from every improvement in any underlying model. This creates a compounding product flywheel, making the application layer an index of the entire AI ecosystem's progress.
The old VC model of taking 30% in a Series A and accepting dilution is being replaced. Now, funds take what ownership the market allows early on and then 'ladder up' to their 20% target by participating in subsequent growth rounds, tenders, and even IPOs. This multi-stage approach is essential for competing in today's market.
Citing a quote from legendary investor Jim Breyer, Miles Clements emphasizes that while the science of VC is valuation, the art is knowing when to ignore it. He shares that Accel missed investing in ServiceTitan, a $9B company, by rigidly adhering to valuation multiples for vertical SaaS, learning a costly lesson about the need for flexibility with generational founders.
Accel Partner Miles Clements reflects on missing Rippling, attributing Parker Conrad's success to his ability to build in downstream revenue levers beyond simple marketing spend. This concept, the 'marginal ease of ARR accumulation,' describes a business's innate ability to compound growth efficiently over time, a key factor investors now seek.
For many high-flying companies from the 2021 peak that are now struggling with growth, the original IPO aspiration is unlikely. Accel's Miles Clements suggests now is a great time to be in the LBO business. Private equity firms will provide the necessary homes for these businesses, offering an exit path different from what founders initially envisioned.
In a moment of candor, Accel's Miles Clements shared that the firm's partners hold global offsites to analyze their biggest misses, such as not investing early in foundation model companies. They grade their performance against the world's top 50 private companies and strategize on how to win the next 50, showcasing a rigorous process for self-correction.
Countering the 'swing for the fences' mentality, Accel co-founder Arthur Patterson's mantra is to 'focus on hitting singles and doubles and let the home runs take care of themselves.' This philosophy encourages investors to focus on fundamentals and strong founder relationships rather than trying to force a $100B outcome from day one, which often leads to failure.