--- title: Ascerta, Formerly Pay-i, Raises $18M to Show Companies Which AI Projects Pay Off description: Ascerta, formerly Pay-i, has raised an $18M Series A led by Dell Technologies Capital to show large companies what their AI projects cost and return. author: Darie Nani (Editor-in-Chief) date: 2026-09-30T14:21:34.995Z updated: 2026-09-30T14:21:35.008Z canonical: https://www.sovereignmagazine.com/article/ascerta-18m-series-a-ai-roi image: https://cdn.nanimediahouse.com/ascerta-co-founders-winters-tepper-holan.webp categories: Artificial Intelligence, Startups content_type: Spotlight region: Washington State publication: Sovereign Magazine about: - type: Organization name: Ascerta description: Ascerta is an Enterprise AI Management company based in Bellevue, Washington. Its platform measures what AI costs an organization, how staff use it and which AI initiatives produce business value. The company was founded in 2024 and was known as Pay-i until September 2026. url: https://ascerta.com industry: Enterprise AI software sameAs: - https://www.linkedin.com/company/ascerta-corp/ --- Ascerta, a Bellevue, Washington company whose software shows large businesses what their AI projects cost and what they return, has raised $18 million in a Series A round led by Dell Technologies Capital. Hitachi Ventures, BGV, Wipro Ventures and earlier investors also took part, bringing Ascerta's total funding to $22.9 million. The company was called Pay-i until this month. It will use the money to expand its platform and its sales and marketing team, and to connect the platform to more of the AI tools that large companies use. Ascerta calls its field Enterprise AI Management. Chief information officers, chief financial officers and AI leaders use its platform to see in one place what AI costs across the organization, how staff use it and which projects produce measurable results. Across its customers, the company says, the platform has improved ROI on AI initiatives by 47%, cut the time taken to launch an agent by 24% and reduced wasted AI spend by 86%. ## Measuring AI ROI in Large Companies Gartner forecasts that [worldwide spending on AI](https://www.sovereignmagazine.com/article/ai-capex-oil-gas-compute-futures-2026) will reach $2.7 trillion in 2026, up 49.5% from last year. In a [Gartner survey](https://www.gartner.com/en/newsroom/press-releases/gartner-survey-finds-only-22-percent-of-organizations-have-successfully-scaled-ai-across-multiple-business-units) published on September 1, which drew 1,303 respondents from organizations with annual revenue of at least $50 million, roughly 11% did not know what their function had spent on AI in 2025, and only 22% of organizations had scaled AI across several business units or adopted an AI-first approach. The organizations that tracked the ROI of their AI initiatives constantly, and stopped or reallocated the weak ones, reported positive returns on 81% of them. > "Companies are counting tokens, lines of generated code, and agent runs, struggling to derive the impact AI has on their business." > — David Tepper "The market is full of meaningless vanity metrics," said David Tepper, Ascerta's co-founder and chief executive. "We built Ascerta to cut through the noise and give organizations the means to win in the AI-era," he said. "That means insights specific to their business, people, and use cases. That means purpose-built tools to prevent waste and aggressively optimize for value." ## Falling Token Costs and Rising AI Spend Tepper told the research firm Forrester in 2025 that his company had tracked model prices since 2022, and that the compute cost of running a model of a given size was falling by roughly 6.67% a month, compounded, while organizations' spending on those models rose far faster. Companies now send much more text to the model with each request, he said, so a single request can cost more than it did a few years ago. [Gartner forecast in March](https://www.gartner.com/en/newsroom/press-releases/2026-03-25-gartner-predicts-that-by-2030-performing-inference-on-an-llm-with-1-trillion-parameters-will-cost-genai-providers-over-90-percent-less-than-in-2025) that by 2030, running a large language model with one trillion parameters will cost AI providers more than 90% less than it did in 2025. It also said agentic models need [between five and 30 times more tokens per task](https://www.sovereignmagazine.com/article/agentic-ai-pay-per-action-pricing-bill-shock) than a standard chatbot, and that overall inference costs would rise because token consumption is growing faster than token prices are falling. Uber president and chief operating officer Andrew Macdonald said on the Rapid Response podcast earlier this year that the company could not yet connect its growing use of the coding agent Claude Code to the number of useful features it ships. "That link is not there yet," he said. His comments followed reports that Uber had used its 2026 budget for AI coding tools within four months. Microsoft has since [set AI token budgets](https://www.sovereignmagazine.com/article/microsoft-ai-token-budget-copilot-engineers) for its engineering divisions. Ascerta says traditional FinOps tools, which track cloud and software bills, can show what AI costs but not what it does for the business. In the FinOps Foundation's 2026 survey, 98% of the 1,192 respondents said they managed AI spending, up from 31% two years earlier, and FinOps for AI was their top priority for the future. ## Ascerta's Atlas, Forge and Convoy Products Ascerta connects to the AI a company already runs, including applications built in-house, Microsoft's Copilot suite, Amazon Bedrock AgentCore, Salesforce Agentforce and coding agents such as GitHub Copilot, Claude Code and Codex. It records what each use of AI costs by tying individual model calls to a specific use case, and it counts sub-token costs, hidden fees and enterprise discounts. The company says no other tool on the market measures AI cost in as much detail. Each use case is then matched to the business KPIs it was meant to improve, so managers can see which initiatives create value, which need fixing and which should be stopped. The platform also tracks adoption by person, team and tool, so an organization can see who is getting results from AI and help everyone else catch up. Of the platform's three products, Atlas measures AI value, adoption and ROI, from a single workflow to a company's whole portfolio of AI projects. Forge shows engineering managers how their teams use coding agents and whether that use is making them more productive. Convoy is for organizations that buy their own AI capacity, helping them use all of it and add new use cases without disrupting the ones already in production. Customers use Ascerta to put dollar values on AI-powered features and to recover money lost to failed agent runs, duplicate projects and shadow AI, the AI tools that staff use without the company's approval. ## Atos and Wipro Among Ascerta's Customers Ascerta's customers include several global insurance carriers and the IT services groups Atos and Wipro, the second of which is also an investor through Wipro Ventures. The company works with partners including Microsoft, AWS, IBM, Slalom and Trace3. Atos launched its Sovereign Agentic Studios in March 2026 to help organizations move agentic AI from pilots into production, and named Pay-i as one of six companies in the Scaler accelerator program that supports them. "At Atos, our Sovereign Agentic Studios operating model is built on moving agentic AI from pilot to production at global scale," said Florin Rotar, Group CTO at Atos. "That requires measurable business value, not just technical capability. Ascerta has been instrumental in delivering on that promise, giving us the visibility and control we need to scale AI initiatives with confidence." ## Ascerta's Founders and Its Start as Pay-i Tepper founded the company in 2024 with Doron Holan, its chief technology officer, and Erik Winters, its chief operating officer. Tepper spent 19 years at Microsoft, where he led generative AI strategy for internal use across Azure. Holan spent 27 years there and designed Azure's throttling infrastructure, which handles hundreds of billions of requests a day. Winters built his career leading growth and go-to-market strategy at early-stage companies, along with the forecasting and analytics behind those decisions. The company came out of stealth as Pay-i in May 2025 with a $4.9 million seed round co-led by Fuse Partners and Tola Capital, selling software for managing AI costs. As its customers adopted AI more widely, they also wanted to know how it was being used, what their agents and models were doing and whether the results justified more investment. Ascerta says that wider brief is the reason for the new name. ## Dell Technologies Capital and Plans for the Series A "Ascerta is building the system of record for AI value creation," said Raman Khanna, managing director at Dell Technologies Capital. "Most enterprises are moving beyond broad AI experimentation and focusing their investments on what delivers measurable business value. David and the Ascerta team are giving leaders the visibility and rigor they need to understand what's working, optimize spending and scale their most successful AI initiatives." Ascerta's integrations already cover nearly every major enterprise AI tool, and the company plans to extend them to all of them. It is also turning its research on AI value into new products that are meant to improve the return on AI as well as measure it. **About Ascerta** Ascerta is an Enterprise AI Management company based in Bellevue, Washington. Its platform measures what AI costs an organization, how staff use it and which AI initiatives produce business value. The company was founded in 2024 and was known as Pay-i until September 2026. [Website](https://ascerta.com) ## FAQ **Q: How do you measure ROI on AI?** By setting what an AI project costs against the value of what it produces. Ascerta ties every model call to a specific use case, then matches that use case to the business KPI it was meant to improve, such as dollars earned or time saved. In Gartner's 2026 survey, organizations that tracked ROI constantly and stopped their weak projects reported positive returns on 81% of their AI initiatives. **Q: What is AI cost management?** AI cost management is the tracking and control of what a company spends on AI models, agents and the capacity to run them. Ascerta started in this field as Pay-i in 2025. It now records costs down to individual model calls, including sub-token costs, hidden fees and enterprise discounts, and sets them against the value each use case produces. **Q: What is FinOps for AI?** FinOps is the practice of managing technology spending, first developed for cloud bills, and FinOps for AI applies it to models, tokens and GPUs. In the FinOps Foundation's 2026 survey, 98% of respondents said they managed AI spending, up from 31% two years earlier. Ascerta says FinOps tools show what AI costs but not what the business gets for it. **Q: What are examples of shadow AI?** Shadow AI is any AI tool used inside a company without its approval, such as an employee using a personal chatbot account for work or a team running a coding agent that the IT department does not know about. Ascerta's customers use it to recover spending lost to shadow AI, failed agent runs and duplicate projects, and the company says its platform has cut wasted AI spend by 86% across its customers.