Cyber models and third-party datasets can provide valuable tools for understanding, quantifying, and managing cyber risks. They enable companies make data-driven decisions, help enhance cybersecurity posture, and empower organizations to effectively respond to the evolving threat landscape. No cyber datasets are one-size-fits-all, nor are they created equally, however. We'll discuss how the world of cyber insurance has unique resources to separate the good from the bad from the snake oil. By using the set of companies that had cyber insurance claims, along with the universe of companies that purchased cyber insurance, we can determine which of these tools has actual correlations with cyber incidents. We'll evaluate cyber self-assessments, outside-in scans, and dark web threat intelligence, amongst others.