What is fingerprint confidence?

When a fingerprint is generated, it is assigned a confidence value. The confidence value is a number between 0 and 1 that indicates the confidence of the determined fingerprint.

Confidence calculation

The confidence value is calculated based on the similarity between the current browser state and the previous state. The components used to calculate these states are called signals. If the browser signals have not changed, the confidence value will be 1. If the browser signals have changed, the confidence value will be lower.

Signal drift

Signal drift is the change in the browser state over time. This can happen for various reasons, such as the browser updating, the user installing or uninstalling extensions, or the user changing their browser settings.

Managing drift

Rupt will automatically manage drift for you. When drift happens, the confidence value will decrease but the fingerprint will remain the same. If the drift is large, the fingerprint will change and a new fingerprint will be assigned.

To improve the stability of your fingerprints, you should call the getFingerprint function as often as possible. This will ensure that the browser state is refreshed and the confidence value is updated.

Erring on the side of caution

Rupt always errs on the side of caution. If the confidence value is low, Rupt will not assign a fingerprint. This is to prevent false positives and ensure that the fingerprint is as stable as possible.

Optimization

Rupt's fingerprints are optimized for SaaS and web applications to prevent fraud draining revenue. This allows Rupt to hone in on the most relevant signals and provide a more accurate fingerprint for these scenarios.