Moving Beyond ‘Black Box’ Scores: How Owlin’s Explainable Scores Transforms Media Monitoring

In this edition of Owlin Solution Talks, we speak with Sjoerd Leemhuis, CEO and co-founder of Owlin, about the importance of explainable scores when monitoring adverse media. Sjoerd tells us why Owlin is committed to transparency in its scoring system, how explainability impacts client confidence, and how Owlin’s unique approach helps businesses make better-informed risk decisions in a world driven by data and regulatory demands.

Why is explainability essential when monitoring adverse media and scoring risk, and how does Owlin differentiate from “black box” scores in this area?

“To understand why explainability is crucial, let’s start with the purpose of scores themselves. At Owlin, we analyze vast amounts of unstructured data like adverse media, often far more than a person could read in a day. Scores help us condense that information, similar to how credit ratings simplify financial assessments. 

Scores make it easy to identify trends and spot when you need to take action—especially in an extensive portfolio where it’s essential to see where risks are highest quickly. But to act on these scores, you need to trust them, and trust comes from understanding what’s behind them.”

A “black box” score—a rating without clear rationale—leaves our clients in a difficult position, who often manage extensive portfolios for risk exposure to third parties. If a score says there’s a risk, but you don’t know why, it’s a problem. You’re essentially asked to take action without knowing the cause or context of that risk. Owlin’s clients need more than that; they need to understand why a score is high or low to make decisions confidently.”

Another challenge our clients face is the increasing regulatory pressure due to regulations like the German Supply Chain Act in Europe or the California Transparency in Supply Chains Act. These regulations require them to monitor third parties actively and provide evidence of follow-up actions. 

Therefore, we often ask how the Owlin tool can help achieve compliance. One of our tool’s core strengths is its explainability. We enable users to document and verify actions based on Owlin’s insights through audit trails, case tracking, and notes. This means that we don’t just provide adverse media monitoring but also support clients in documenting how they’ve addressed identified risks. 

How does Owlin make its scores transparent and trustworthy?

“Owlin’s commitment to transparency begins with understanding that risk scores don’t need to explain every minor detail—they need to clarify the main factors that drive each score. This approach ensures that users can quickly see the underlying components (scale) without exhaustive explanations. 

That said, our scores are structured around specific categories of risk events that we openly document. These include fraud, strikes, corruption, data breaches, and other impactful factors. Each score is tied directly to the actual risk events in the news about a company or entity, helping clients quickly recognize the existence and cause of a risk.

Owlin categorizes risks and helps clients create a ‘blueprint of the news’ through our Stories feature. This feature provides context around each event, connecting related articles and news developments to view the situation comprehensively. It gives users the broader story behind risk events, making understanding their impact and timeline easier.

To ensure scores are actionable, Owlin shares insights into the methodology behind them. Our machine learning models identify and categorize risk events and then enrich the scores with explanations that clarify why each score applies. Owlin allows access to original sources—news articles and reports—supporting each score for clients who wish to investigate further.

Clients can also customize scores to match their specific risk priorities. For instance, if data breaches are a high priority, clients can adjust filters to give more weight to cybersecurity news, tailoring their portfolio view to align with their unique risk landscape.

How is Owlin planning to enhance the explainability of scores as AI continues to evolve?

“We continuously look for ways to improve our product and leverage AI more innovatively. A big drive for us to do this is that we’ve noticed that our clients need to manage more risk with the same team due to increasing demands from the market, regulations, and company leadership. 

Owlin’s AI-powered tools help our clients stay ahead by making identifying risks quickly and efficiently easier. However, we know that the quality of the output from AI models is only as good as the data we feed them, which is why we’re always focused on improving data quality and expanding our data sources.”

Thank you, Sjoerd!

Meet Other Owls

• Stefan Peekel about Owlin’s Approach to Adverse Media Monitoring
• Eduardo about being a Product and Customer Success Manager at Owlin
• Physicist Willem Westra about how he applies physics in his daily work for Owlin
• Ralf Nieuwenhuizen about how Owlin can analyze a million articles in real-time
• Judith Landstra – de Graaf about Marketing for Owlin
• Lars Kuijlenburg about Software Engineering at Owlin