Many startup founders know all too well the daunting feeling of walking into a venture capital firm, preparing to give the pitch that could change their company’s future.
Sand Hill Road, which winds through famous Silicon Valley towns such as Menlo Park and Palo Alto, has become synonymous with the modern startup boom, as it is home to many of the industry’s most prominent VC firms.
💰💸 Don’t miss the move: SIGN UP for TheStreet’s FREE Daily newsletter 💰💸
For years, the leaders of fast-growing startups have battled for the chance to spend just a few minutes in the board rooms of firms such as Sequoia Capital or Redpoint Ventures. Now, the process of getting in the door may be about to get significantly harder.
One of the industry’s most prominent data providers has announced a new artificial intelligence (AI) tool that could change the ways that VCs evaluate startups. But some experts advise investors to approach it with caution.
A leading startup data provider is making an important change
If you’ve ever looked up information on a startup, either early or late-stage, you probably turned to Crunchbase. The platform is known for providing key data on many companies, from funding history to updates to leadership changes, often serving as a convenient source for market research and industry trend analysis.
Now Crunchbase is making a major change, rolling out a new tool that could significantly impact the startup and venture capitalist communities. Specifically, it is going to become a prediction engine, powered by AI to forecast not a startup’s growth trajectory but also funding rounds and acquisitions.
Related: AI startup smashes funding round, signals big changes for healthcare
In its announcement, Crunchbase made strong proclamations, stating that “historical data is dead” and that the most valuable data can help make informed future predictions. It even claims that its new AI engine can forecast startup events at a 95% accuracy rate.
This raises an immediate question: how does Crunchbase’s new AI predict startup events with so much accuracy? VentureBeat recently provided context on how it works, stating:
“Instead of focusing solely on past events, Crunchbase now leverages its massive dataset — including usage patterns from 80 million active users — to predict future business outcomes. The company’s AI analyzes thousands of signals to forecast events.”
In a world where historical data is no longer enough, we’re introducing the ✨ NEW ✨ Crunchbase: AI-powered predictive intelligence. We predict private market movements, from funding to acquisitions, and identify the companies set up for success. 🔮
Explore more at… pic.twitter.com/LDARSMkYhN
— crunchbase (@crunchbase) February 19, 2025
Eric Vaughan, CEO of IgniteTech tells TheStreet that the ways in which venture capitalists define value has evolved to the point that metrics such as revenue growth and profitability are simply the foundation. He highlights a “complex matrix of factors” that VCs consider, including customer metrics, market position and capital efficiency.
AI startup founder and processor of data science Kathryn Wifvat adds that many VCs would describe a successful startup as one that can achieve an exit through acquisition or initial public offering (IPO ) within a 5-10 year window, “delivering at least a 10x return on their initial investment.”
- Elizabeth Holmes shocks the world with first interview from prison
- OpenAI rival powers toward milestone with massive implications
- Dating app company makes shocking change to enhance user safety
The principles of a successful startup, at least according to a venture capitalist, appear fairly straightforward. But that doesn’t mean that this model is a perfect means of evaluating them.
Experts are skeptical that the 95% success rate is accurate
As Wifvat sees it, assessing the model’s accuracy is difficult without knowing the exact data used to train it. “Startup success is influenced by unpredictable factors like policy changes and market shifts,” she notes, highlighting another potential complication.
She isn’t the only expert with a skeptical take on Crunchbase’s claims. Komninos Chatzipapas, founder of HeraHaven.AI, expresses a similar outlook, noting that it is extremely difficult to predict future business events, as its equations often require internal business data.
Related: OpenAI considers taking drastic action to stop Elon Musk
He adds that “connecting the dots looking backwards” is often fairly straightforward once the necessary data becomes public after the event. “There is no shortage of algorithmic trading strategies that achieve amazing results in backtesting, but fail terribly when put live because they’ve overfitted on the old data,” he states.
Michael Ashley Schulman, Partner and Chief Investment Officer at Running Point Capital Advisors questions replicability backtested 95% AI accuracy rate. “We know that Crunchbase continuously adds data to its files but a prediction needs to be made at a single point in time with the information at that moment,” he states.
Angel investor Kevin Korte sees CrunchBase’s definition of success as misleading.
He also raises other red flags, stating, “They are silent on the number of startups the AI declined to invest in, yet they managed to raise money. If I miss 100 good investment opportunities because the AI sees them as unworthy, I miss them regardless of whether I correctly predict the remainder.”
“They are silent on the number of startups the AI declined to invest in, yet they managed to raise money. If I miss 100 good investment opportunities because the AI sees them as unworthy, I miss them regardless of whether I correctly predict the remainder.”
Even those who don’t see the 95% figure as being inaccurate, note that it poses limitations for evaluating startups. Elliott Parker, CEO of High Alpha Innovation, notes that the Crunchbase model likely won’t be able to assess the magnitude of a startup’s success.
“It will be great at determining if a company is getting close to raising another funding round” he states, “but it won’t be able to predict the valuation, terms of the round, the lead investor, etc. that determine “success.”
Related: Veteran fund manager issues dire S&P 500 warning for 2025