Startup Ascerta collects $18M in funding to help enterprises understand the value AI provides

Ascerta has raised $18 million to help companies answer a question a lot of AI budgets still dodge: what, exactly, is this buying us? The early-stage round was led by Dell Technologies Capital, with Hitachi Ventures, BGV and Wipro Ventures also taking part. That lifts the startup’s total funding to $22.9 million.

The pitch is simple, and a little uncomfortable for anyone signing off on big AI spend. Businesses are getting very good at tracking token counts, agent runs and lines of code. They are much worse at connecting those activity stats to revenue, savings or productivity. David Tepper, Ascerta’s co-founder and chief executive, says the problem is that traditional FinOps-style tools were built for cloud spending, not for the business outcomes of AI agents and coding assistants.

Ascerta’s platform tries to fill that gap with a single system that shows how AI is being used, what it is doing and whether it is producing money or savings. It connects to enterprise tools including Microsoft Corp.’s CoConvoypilot, Anthropic PBC’s Claude, Amazon Web Services Inc.’s Bedrock and Salesforce Inc.’s Agentforce. Once plugged in, it tracks costs down to users, teams and applications, and Tepper says it can also expose hidden fees, enterprise discounts and sub-token costs that other tools miss.

The product is split into three parts. Atlas measures adoption, value and return on investment. Forge looks at how engineering teams use coding agents and how that affects productivity. Convoy tracks compute use so companies can consolidate workloads and cut waste without breaking operations.

Ascerta says customers including Atos SE and Wipro Ltd. have already used the platform to improve AI ROI by an average of 47%, reduce agent launch times by 24% and cut wasted AI resource spending by 86%. Atos Group Chief AI Officer Florin Ratar called it instrumental to scaling the company’s Sovereign Agentic Studios initiative. Dell Technologies Capital’s Ramana Khanna framed the bet as backing a system of record for AI value creation, not just another dashboard full of vanity metrics.