Signal Uniqueness
Signal uniqueness is a measure of how distinctive a signal is, reflecting how much identifying information it carries about a device. Highly unique signals such as canvas or WebGL output contribute far more to identification than common ones like platform or language that most users share.
How Signal Uniqueness works
Uniqueness depends on how a signal's values are distributed across the population. If almost everyone reports the same value, the signal barely narrows down who a device is; if values are spread across many small groups, knowing the value strongly distinguishes a device. This is closely related to entropy.
Fingerprinting systems estimate uniqueness empirically from observed distributions, then weight each signal accordingly. Distinctive signals like rendering-based canvas and WebGL hashes carry heavy weight, while broadly shared signals like operating-system platform or preferred language carry little on their own but still add corroboration.
Uniqueness must be balanced against stability, because the most distinctive signals are sometimes the most volatile or the most likely to be perturbed by privacy tools. Good identification blends highly unique signals with stable ones so the combined profile is both distinctive and durable.
Why Signal Uniqueness matters for fraud prevention
Signal uniqueness governs how confidently a device can be told apart from others, which directly shapes the reliability of fraud detection. Weighting signals by uniqueness prevents common attributes from diluting a match and ensures distinctive evidence drives recognition. It is central to producing well-calibrated confidence in whether two sessions belong to the same device.
How TRACIO handles it
TRACIO weights its 130+ signals by their uniqueness so that distinctive rendering and hardware signals count more than commonly shared attributes when forming a visitor ID. Balancing uniqueness with stability lets TRACIO build profiles that are both distinctive and durable, supporting its identification accuracy on internal benchmarks. This weighting is a core part of how the confidence in a match is calculated.
Related terms
Explore further
Frequently asked questions
Identify every device with confidence
Start with a free plan of 2,500 API calls per month. No credit card required.