SaaS Sessions · 2025-09-12 · 37 min
Every SaaS company is racing to “add AI,” but most are doing it wrong. In this episode, Megh Gautam, Former Chief Product Officer at Crunchbase, reveals the hard truths behind building AI into established SaaS products. From avoiding hype-driven features to building trust through data quality and transparency, Megh shares how Crunchbase rolled out AI-powered capabilities without breaking user trust. He also breaks down the internal alignment, cross-functional execution, and relentless feedback loops required to ship AI features that actually matter. Key Takeaways - Start with Real User Problems AI should not be an “add-on story” - it must solve a core customer pain. Crunchbase began with AI in search, a high-usage, high-friction feature. Prioritize critical workflows over “nice-to-have” gimmicks. Data Quality Determines Trust Bad data in = garbage out, especially with AI models. Crunchbase spent a decade building clean, reliable data pipelines before layering AI. Trustworthy results require grounding AI outputs in verified “truth sets.” User Trust Demands Transparency Customers don’t just want answers - they want to know how those answers were derived.
S10E2 - From Harvard Law to SaaS CEO: Decoding the "Paperless" Future ft Shashank Bijapur, Spotdraft65 / 100
S10E1 - Engineering the Super Bowl ft. Catherine Johnson, VP of Global Solutions Engg. at Hydrolix
S9E8 - Unified Commerce & AI ft. Jigar Dafda, CPTO at Fynd
S9E7 - From Clicks to Conversations ft. Tom Drummond, Founder & Managing Director at Heavybit Industries
S9E6 - The Future of Intelligent Agents in SaaS ft. Bhaskar Roy, Chief of AI Products and Solutions at Workato