Larridin $30M Series A: Agent ROI Measurement Research (Nov 2026)

## Key Findings (Nov 5, 2026) ### Funding (Verified) - Larridin raised $30M Series A on Nov 2, 2026, led by Pelion Ventures. Existing investors: a16z, Gradient Ventures. New: GFT Ventures, W23, Thayer, DG Daiwa Ventures, Alumni Ventures, Cougar Capital. Total funding: $47M (including $17M seed). Sources: larridin.com press release, siliconangle.com, siliconreport.com ### Product Claims (Company-Reported, Unverified) - AI value measurement platform with 4 modules: Spend Intelligence, AI Impact, Developer Intelligence, Workflow Intelligence. - Claims 47% average increase in measurable team output, median $2.4M AI spend "optimized" in year one, 2B+ sessions measured, 82K+ workflows tracked. - Named customers: SurveyMonkey, Vertiv, Klaviyo, Gainsight, Globality, TigerConnect, EcoVadis, Sundt. Company says 50+ customer organizations. - Founders: Russ Fradin (CEO), Jim Larrison, Ameya Kanitkar (CTO) — former Comscore executives. ### Broader Agent ROI Problem (Independent Surveys) - **KPMG Global AI Pulse (n=2,145, Apr-May 2026):** 42% of leaders have only partial visibility into AI spending; 23% struggle with usage-based costs. Organizations with cost-monitoring dashboards were 5x more likely to report established ROI (15% vs 3%). Source: kpmg.com - **Dataiku/Harris Poll CIO survey (n=685):** Only 24% of CIOs have mature ROI measurement across most AI initiatives. Only 21% have near-real-time visibility into agent/advanced AI workload costs. 72% don't consistently measure both operational performance and business outcomes for every agent. Source: dataiku.com - **Mavvrik/Benchmarkit (n=396):** Only 11% forecast AI spend within ±10%. Cost reporting coverage is 36% for agentic workflows vs 67% for public cloud. Source: mavvrik.ai - **WitnessAI/Harris Poll (n=300):** Only 9% say more than three-quarters of AI initiatives have delivered measurable financial return. Source: witness.ai ### Caveats - Larridin's outcome metrics are self-reported with no published methodology. - Privacy claims about prompt handling conflict between sources (Business Insider says it analyzes employee prompts; SiliconReport says it evaluates interactions without storing prompt contents). - No independent customer confirmations beyond company's own list.