Product review
Castle device fingerprinting review
Castle combines client and server device evidence with account activity. Its APIs return separate account-abuse, account-takeover and bot scores; policies and custom signals let the application adapt those outputs to its own workflows.
Key facts
- Device intelligence
- Combines browser/app and server context with historical devices and locations. The documentation describes device fingerprinting and links to account-abuse use cases. Source ↗
- Scores
- Risk and Filter APIs compute separate Abuse, ATO and Bot scores. API values run from 0.0 to 1.0, represented as 0–100 in the dashboard. Source ↗
- Free and Pro
- Free includes 5 US dollars of monthly API usage, equivalent to 1,000 Risk calls or 5,000 IP lookups at published rates. Pro includes 200 US dollars of usage, equivalent to 40,000 Risk calls. Source ↗
- Operations
- Plans differ in request rates, event retention, environments and support. Enterprise starts at 4,000 US dollars per month with custom volume/MTU options. Source ↗
Strengths
- Ready-made risk scores alongside device intelligence.
- Developer documentation, policies, webhooks and transparent usage units.
Limitations
- Count the mix of API types against the included credit.
- Short retention and request-rate limits on entry plans may affect investigation and peak traffic.
Who it fits
Developers combining device identification and ready-made abuse scores with product-specific policies and event workflows.
Compare the alternatives by identity scope when your team needs a different identity product or operating model.
What the score means
The editorial score of 8.4 out of 10 places this tool in the published comparison. It reflects product fit under the same 42-criterion framework used for all ten tools.
The facts above help evaluate identity persistence, tamper resistance, risk context, API experience, production coverage, evidence and commercial fit. Read every criterion and weight.
How to evaluate it
Join client collection to a Risk API event on signup or login. Inspect all three scores, the device relationship and policy output using a legitimate-user and known-test matrix.
Self-serve evaluation account
The Free plan is an account-based evaluation route with published limits, not an anonymous browser checker.
Open evaluation routeUse the public route to inspect access and output, then validate the product on your own permitted traffic.
Sources and further reading
- Product and detection approach · checked October 21, 2026
- Pricing and access · checked October 21, 2026
- Integration and result documentation · checked October 21, 2026
- Official technical guide and further reading · checked October 21, 2026
FAQ
Does Castle require building every risk model from scratch?
No. Castle returns separate Abuse, ATO and Bot scores. Policies and custom signals configure how those results fit the business workflow.
Are Risk API calls and IP lookups priced the same way?
No. The pricing page lists different rates and expresses plan access as an API-usage credit. Estimate the actual mix of request types.
Explore checkers, demos and evaluation routes →