In May 2022, investors in San Francisco-based 3D graphics platform Unity Technologies were blindsided. The company’s share price fell 39% in a week, not because demand had slowed, but because a machine learning model had been fed bad quality information, costing $110m in potential revenue and wiping out $10bn of Unity’s market cap.

Today, the risk surface of bad quality data is even greater than it was then. As enterprises rapidly adopt AI, with agents taking actions off the back of company data, bad inputs lead to bad decisions that, on top of lost revenue, can result in breaches of regulations, huge fines and reputational damage.

This is pouring gasoline on the market for enterprise data management, which is growing from $125bn in 2025 to a predicted $350bn by 2034, but legacy products in the space are often slow, unreliable and don’t give enterprises the tools they need to implement robust controls.

Stockholm-based Validio develops an enterprise-grade platform to detect and fix data issues quickly and reliably. The company has seen 8x revenue growth in the past year, signing up huge customers who say the product vastly outperforms alternatives on the market.

Validio is led by a remarkable founder in Patrik Liu Tran. With a PhD and 15 years’ worth of academic experience in business and AI, he combines technical credibility with proven commercial flair and deep experience of how enterprises manage sensitive information, having built more than 20 in-house data quality products for big companies.

This round will allow the company to move even faster, as it builds the category-leading platform to help enterprises make their data AI-ready.

Becoming a no-brainer

Speaking to Validio’s customers illustrates just how broken the current market for this kind of tool is.

“I spent months setting up this other expensive tool, it was slow, it missed problems, and generated false positives,” they will say. “Then I tried Validio, it was set up in a day and added value immediately.”

Customers have even signed up to Validio before their previous contract with a more established competitor has expired, and the company has so far seen zero churn among enterprise customers.

That kind of product-market-fit has seen the company sign $1m+ contracts with customers who range from established tech companies like Canva and Surfshark, to leading banks and financial institutions like Nordea and AllianceBernstein.

Pressure to automate fast is intensifying demand for Validio’s product, as enterprises realise that AI is only going to be as good as the information you feed it. 95% of AI initiatives fail to reach production, with one of the main reasons being poor data quality.

This compounds the existing need for large organisations to maintain a high-quality, secure data environment, be it in highly-regulated industries like healthcare and financial services, or Fortune 500 companies holding large quantities of sensitive customer information.

Best-in-class tech

Validio has built its product to work for these kinds of businesses where data privacy and security concerns really matter, by allowing customers to run it on their own secure server, something that few competitors offer.

The tool also outperforms alternatives due to its ability to assess enterprises’ actual data, rather than just metadata. This means it can automatically digest years of historical information to train the AI to detect anomalies faster, and with more detail and context.

Where other tools assessing metadata might be able to tell you: “Something is broken because this table is no longer feeding into that table”, Validio will tell you exactly what is going wrong: “Your revenue metric from last week cannot be right, as the average order size has gone up, but revenue didn’t.”

Validio understands what data is available, its quality, where it is coming from and where it is heading.

Having this granular context allows the platform’s AI agents to recommend, and implement, data quality controls automatically, a job that has historically been highly manual, labour-intensive and expensive. As issues are detected, the AI agents notify the person responsible for managing or using that data, and help identify the root cause, with an interface that makes sense to both technical and non-technical team members.

The platform’s deep observability of data and ability to quickly take action off the back of it is possible due to Patrik’s decision to build it on high-performance Rust architecture. Most legacy alternatives are built on less performant coding frameworks such as Python, which processes actions 50-100x slower than Rust.

This is an example of first principles thinking and conviction in technical expertise informing a contrarian bet that is now really paying off.

In 2019 when Patrik chose Rust, many dismissed the programming language as hype. Today, Elon Musk is describing it as “the language of AI” with developers preferring it over Python due to its speed and the size of datasets it can handle.

Ambition backed by action

This kind of self belief comes from a founder who has shown serious grit and ambition to get where he is today.

Patrik grew up in a small town in southern Sweden best known for high rates of crime, drug use and gang violence. He saw that, for himself and his friends around him, there were essentially two options: you either get involved or you get out.

He remembers watching a Swedish-noir crime movie, Easy Money, which moves between the country’s tech scene and its criminal underworld. While his friends saw the film and were attracted to the promise of easy money to be made from running drugs, Patrik was inspired by the main character who studied at Stockholm School of Economics and had a promising alternative future. So Patrik travelled up to Stockholm School of Economics by train to see if that alternative future was real, and when he saw that everyone there could afford MacBooks and nice clothes, he began working as hard as he could to make that his reality.

He graduated from high school early at 16 years old, and began a double degree in parallel, in business at the Stockholm School of Economics and machine learning at KTH Royal Institute of Technology.

Halfway through, after receiving internship offers to crunch numbers at investment banks, which he already knew he could do, he decided to use the summer to learn how to sell.

He talked his way into a role at a Swedish high street bank, promising them he would up their revenues by cold calling. When the summer was over, he was so successful that he won the bank’s national sales competition for being the most successful cold caller.

Patrik also understands the problem that he’s solving for Validio’s customers intimately. While completing his PhD, he spent five years as a consultant advising on AI and data strategy for more than 30 enterprises and top European companies.

Patrik is the kind of founder that leading angel investors chase, with one top Swedish entrepreneur offering him a €500k cheque, regardless of what company he ended up building. He has also built a strong team around him, with hires that include a Harvard PhD, and senior operators with experience at the likes of Spotify, AWS and EF.

Category leadership

With Validio, Patrik has built a product that serves a mission-critical business function for enterprises, which customers say is better than any other tool on the market.

Aside from helping companies stay regulation-compliant and cut costly risks, it will unlock huge value for users, improving outputs from powerful new AI by allowing them to sanity-check the quality of the inputs.

Validio’s elegance, ease of use and industry-leading performance make it a very difficult platform to stop using. This new round will give the company the firepower it needs to cement its position as the category-leading data tool to help enterprises generate value with AI.