Larridin raised $30 million in Series A funding led by Pelion Ventures, bringing total financing to $47 million. The company measures how enterprise spending on licenses, tokens and agents converts into completed work. It reports more than 2 billion AI sessions measured and over 82,000 workflows tracked. For executives, the pitch is evidence of return on AI investment rather than adoption counts.
How Larridin measures AI work
The November 2 announcement says the capital will support product development and team expansion. Existing investors Andreessen Horowitz and Gradient Ventures participated alongside GFT, W23, Thayer, DG Daiwa Ventures, Alumni Ventures and Cougar Capital. Leadership includes co-founder and CEO Russ Fradin, co-founder and president Jim Larrison, and co-founder and CTO Ameya Kanitkar. Fradin and Larrison previously built digital measurement at Comscore. The stated aim is a comparable measurement layer for work performed with AI.
The AI Impact product combines invoiced licenses, token usage and cloud-model costs with observed workflows and engineering-system data. Spend is mapped to departments and compared with work produced against that investment. The methodology separates AI-supported capacity from realized cash savings. Human-equivalent hours can estimate AI contribution, but their financial value depends on the baseline and how freed capacity gets used. More available time does not automatically cut budgets or staffing.
The Workflow Intelligence platform uses a browser extension and desktop agent to observe activity in applications selected by administrators. It collects metadata such as active applications, their sequence, time spent in each step and transitions between tools. Message bodies, document contents, screen recordings and keystrokes are not collected. Views aggregate at team level by default, with individual views disabled unless administrators enable them. Observations group into workflows to examine duration, repetition, friction and AI involvement.
What measurement means for AI budgets
For companies buying AI, the distinction between adoption and value changes budget conversations. Seat counts and token consumption show usage, while Larridin links usage to a useful result and its production cost. The platform ranks automation opportunities and tracks them through pilots and completed deployments against captured baselines. A workflow flagged as a candidate has not yet delivered the same benefit as one running in production. Small firms get a clearer basis for renewing licenses, while large enterprises can compare departments on common cost and output terms.
Engineering measurement adds quality and review to the picture. Developer Intelligence connects output with AI cost, delivery speed and code quality, including output per $1,000 of AI spend, defect rates and pull-request cycle time against a team baseline. It examines agent sessions, captured steps, token usage and estimated cost, with model routing to set cost and quality priorities. The Q3 2026 engineering report found AI-assisted share of merged pull requests rose from 35% in April to 62% in September, while median waits for initial human review rose 50%. Generating more code does not guarantee faster delivery if review and rework lag.
Whether this approach becomes standard will show in customer retention and expansion of measured baselines over the next funding cycle. Larridin reports a 47% average increase in measurable team output and median first-year AI-spend optimization of $2.4 million across its base. Those are company-reported results, distinct from illustrative dashboard figures with example ROI multiples. If more enterprises publish savings tied to tracked workflows, measurement could shape how AI budgets get approved.
