Leadbay Raises $4.3M to Bring AI Sales Intelligence to Underserved SMBs

What happened

Leadbay, a San Francisco-based AI-powered B2B sales prospecting platform, has raised $4.3 million in seed funding. The round was backed by Y Combinator (Leadbay is part of the YC F25 batch), Rebel Ventures, Roosh Ventures, Inovexus Ventures, TS Ventures, Alumni Ventures, Bright Ventures (the venture arm of Bright Data), Transpose Platform, and Deel Ventures, alongside angel investors from Deel, Gusto, and Pennylane.

The company targets a segment of the B2B market that traditional sales intelligence platforms systematically ignore: small and mid-sized businesses in industries like construction, manufacturing, hospitality, retail, and B2B services — businesses with minimal or no digital footprint. Leadbay claims that platforms like ZoomInfo, Apollo, LinkedIn Sales Navigator, and Clay miss approximately 70 percent of U.S. companies and cover only 0.001 percent of digital signals and intent data for these underserved verticals.

Co-founded by CEO Ludovic Granger and CTO Milan Stankovic (who holds a PhD in AI focused on working with limited data), Leadbay has built a proprietary multi-modal inference model that identifies and qualifies SMB leads that cannot be found through conventional database searches. The company plans to use the funding to expand its U.S. go-to-market team in San Francisco, strengthen its AI inference model through a research partnership with Sorbonne University, and grow its engineering team.

Why it matters for practitioners

Leadbay's funding is notable not for its size — $4.3 million is a modest seed round — but for what it reveals about the structural blind spots in the B2B sales intelligence market and where the next wave of competitive disruption may emerge.

1. The "digitally scarce" segment is economically significant and systematically underserved. The B2B sales intelligence industry has been built around companies that leave digital footprints: website visits, content downloads, LinkedIn engagement, software usage signals. This architecture inherently excludes the vast majority of small and mid-sized businesses in traditional industries. A roofing contractor, a regional manufacturer, or a family-owned restaurant chain does not generate the digital signals that intent data platforms rely on. Yet these businesses collectively represent trillions of dollars in purchasing power and are active buyers of everything from commercial insurance to fleet management software. Leadbay's premise is that AI inference models can qualify these leads without relying on the digital breadcrumbs that existing platforms require — a fundamentally different approach to prospecting.

2. The timing aligns with structural shifts in the sales intelligence market. Leadbay's funding arrives as the dominant players in B2B sales intelligence face their own strategic inflections. ZoomInfo recently cut 600 jobs and is pivoting toward consumption-based pricing and upmarket enterprise accounts. Apollo.io has grown rapidly but remains focused on the digitally active segment. LinkedIn Sales Navigator indexes only against its own network. The result is a market where the incumbents are either retrenching (ZoomInfo), going upmarket (ZoomInfo again), or constrained by their data sources (Apollo, LinkedIn). Leadbay is positioning into the gap that these movements create — not competing head-to-head on the same digitally rich accounts, but pursuing the accounts that no one else can find.

3. The AI inference approach challenges the database paradigm. Traditional sales intelligence operates on a database model: collect data, store it, sell access to it. Leadbay's approach is different — it uses a multi-modal inference model that combines exhaustive market data, internal company knowledge (CRM, ERP, CSV imports), and sales rep instincts to predict which companies are likely to buy. This is closer to a recommendation engine than a database. For market intelligence practitioners tracking the evolution of the sales intelligence category, this distinction matters. If inference models can reliably identify and qualify leads that database models miss, the addressable market for sales intelligence expands dramatically — and the competitive dynamics of the category change with it.

4. The investor mix signals strategic validation. Bright Ventures, the venture arm of Bright Data (a major web data infrastructure provider), participating in the round suggests that the data infrastructure community sees value in Leadbay's approach to extracting intelligence from sparse digital signals. Deel Ventures' participation connects Leadbay to a company that itself sells to SMBs globally. The Sorbonne University research partnership adds an academic dimension to the AI inference model that most sales intelligence startups lack.

Key details

• Company: Leadbay (San Francisco, CA)
• Round: $4.3M Seed
• YC batch: F25
• Lead investors: Y Combinator, Rebel Ventures, Roosh Ventures, Inovexus Ventures, TS Ventures, Alumni Ventures
• Strategic investors: Bright Ventures (Bright Data), Transpose Platform, Deel Ventures
• Angel investors: Founders/executives from Deel, Gusto, Pennylane
• Co-founders: Ludovic Granger (CEO), Milan Stankovic (CTO, PhD in AI)
• Target market: SMBs in construction, manufacturing, hospitality, retail, and B2B services
• Claimed coverage gap: Traditional platforms miss ~70% of U.S. companies
• Technology: Proprietary multi-modal AI inference model for lead identification and qualification
• Research partner: Sorbonne University
• Use of funds: U.S. go-to-market expansion, AI model development, engineering team growth

Market implications

Leadbay's entry into the sales intelligence market illustrates a pattern that is playing out across the broader market intelligence landscape: AI is not just improving existing workflows — it is making entirely new categories of intelligence possible. The businesses that Leadbay targets are not new. Construction contractors, manufacturers, and hospitality operators have always existed and always purchased B2B products. What is new is the ability to identify and qualify them at scale using inference rather than observation, prediction rather than tracking.

For the competitive intelligence tools market, this raises an important question about what constitutes a "competitive intelligence tool" in 2026. The best competitive intelligence tools list has traditionally been dominated by platforms focused on digitally active enterprises — companies with websites, social media presences, and technology stacks that can be tracked. If inference-based intelligence platforms can reliably target the offline-heavy segment, the market definition expands, and the tools landscape must expand with it.

The competitive implications for ZoomInfo and Apollo are indirect but real. Neither company is likely to view Leadbay as an immediate competitive threat — the target markets are different. But if Leadbay demonstrates that AI inference can identify and qualify SMB leads as effectively as database-driven approaches qualify enterprise leads, the model will inevitably be applied upstream. The inference approach does not have a natural ceiling at the SMB segment; it is simply starting there because the need is most acute and the incumbents are most absent.

For sales teams selling into traditional industries — insurance, financial services, logistics, facilities management — Leadbay represents a potentially significant expansion of their addressable pipeline. The standard playbook for these teams has been to work referral networks, attend trade shows, and manually prospect through industry directories. An AI platform that can surface and qualify leads in these verticals without relying on intent data signals that these businesses simply do not generate would represent a meaningful productivity gain.

The Sorbonne University research partnership is worth monitoring as well. Academic-commercial collaborations in AI inference modeling are relatively rare in the sales intelligence space, and the results of this partnership — particularly around making AI work effectively with limited and noisy data — could have applications well beyond Leadbay's immediate use case. If the inference model can be generalized, it may become a foundation for a new generation of intelligence tools that operate in data-sparse environments across multiple industries.

Related resources

• Intent Data — how intent data models work and why they fail for digitally scarce businesses
• Market Intelligence — the expanding definition of market intelligence as AI enables inference-based approaches
• Best Competitive Intelligence Tools — the evolving landscape of CI and sales intelligence tools, including emerging SMB-focused entrants