VCs Stress the Importance of Proprietary Data for AI Companies
AI companies across the globe raised more than $100 billion in venture capital dollars in 2024, an increase of more than 80% compared to 2023, according to Crunchbase data. This massive growth in funding now represents nearly a third of the total VC investments for the year. The influx of investment is fueling a crowded industry landscape.
The AI industry has experienced tremendous expansion over the past two years, leading to a market filled with overlapping companies. Many startups use AI merely as a marketing tool, while genuine innovative companies strive to establish themselves. Investors now face the challenge of identifying startups capable of becoming category leaders. So where do they start?
The Importance of Proprietary Data
A recent survey conducted by TechCrunch among 20 venture capitalists (VCs) focused on enterprises unearthed a key factor for AI startups seeking differentiation: proprietary data quality or rarity. More than half of the VCs surveyed emphasized this aspect.
Paul Drews, managing partner at Salesforce Ventures, commented on the difficulty AI startups face in maintaining a moat due to the rapidly evolving landscape. He said that he seeks companies with a blend of unique data, technical innovation, and compelling user experiences.
Jason Mendel of Battery Ventures echoed similar thoughts: "I'm looking for companies that have deep data and workflow moats. Access to unique, proprietary data enables companies to deliver better products than their competitors."
Scott Beechuk from Norwest Venture Partners pointed out the significance of proprietary data, particularly for companies developing vertical solutions. According to him, startups that leverage their unique data hold the most long-term potential.
Andrew Ferguson of Databricks Ventures noted that rich customer data, which enhances a system's feedback loop, improves effectiveness and helps startups differentiate themselves in the market.
Case Studies and Expert Opinions
Valeria Kogan, CEO of Fermata, a startup utilizing computer vision to detect plant diseases, attributed Fermata's traction to training models on both customer data and in-house R&D data. She highlighted the importance of conducting data labeling internally to enhance model accuracy.
Jonathan Lehr of Work-Bench emphasized the significance of not only possessing data but also effectively cleaning and utilizing it: "We focus on vertical AI opportunities addressing specific business workflows requiring deep domain knowledge."
Beyond data, VCs expressed that they seek AI teams led by strong talent, with robust integrations into existing technology and a deep understanding of customer workflows. These factors contribute to an AI startup's potential for market success.