Frequently asked questions
Answers to common questions about the platform, integration, and compliance.
Platform & Detection
What types of fraud does GuardHive detect?
GuardHive detects card-not-present fraud (CNP), account takeover, synthetic fraud, friendly fraud (return abuse), credential stuffing, card testing, and merchant-initiated fraud. AML screening and sanctions checks are available as an integrated layer.
How does real-time decisioning work?
When a transaction arrives, Device Intelligence, Machine Learning models, and your custom rules run in parallel. Results combine into a single risk score and recommended action (approve, challenge, block) — returned in under 50ms including network latency.
How accurate is the fraud detection?
Machine Learning models typically achieve 85–95% precision on card fraud when tuned to your business. Precision is often more important than recall — you'd rather miss some fraud than block too many good transactions. We optimise to your risk appetite.
What data does GuardHive use to make decisions?
Device fingerprints, IP addresses, geolocation, email reputation, phone validation, card metadata, transaction velocity, behavioural patterns, and your custom rules. You control which signals feed into the final decision.
Can I customise the rules?
Yes. GuardHive includes a full rules engine. Define thresholds, weighted conditions, and decision trees without code. Rules update in real time with no downtime.
What's the difference between Device Intelligence and Machine Learning?
Device Intelligence identifies devices and tracks patterns (same device across sessions, devices linked to fraud rings). Machine Learning models learn from historical data and flag anomalies (unusual behaviour, emerging attack vectors). Together they are stronger than either alone.
Integration & Implementation
How long does integration take?
Most integrations are live in 1–2 weeks. Sandbox testing takes 3–5 days followed by production cutover. Complex implementations (AML, custom case management workflows) add 2–4 weeks. Your dedicated analyst manages the timeline.
What's the API response time?
Sub-50ms at the p99 percentile. Most requests complete in 20–30ms. Network latency is included in the SLA.
Do you offer SDKs?
Yes. We provide REST API documentation and SDKs for Node.js, Python, Java, and Go. Language not listed? Contact us — we can add it.
Can I use GuardHive for batch processing?
Yes. You can send decisions at authorization time, at capture, at settlement, or in batch. The API is the same; you choose when to call it.
What happens if your API is slow or down?
GuardHive runs on a 99.9%+ uptime SLA across all plans, with multi-region failover. If the API is unavailable, you can either fall back to your legacy system or set a default action (e.g. approve all transactions pending retry). We recommend a fallback strategy as part of integration.
Do you support webhooks?
Yes. Subscribe to transaction decisions, case state changes, model updates, and alert events. Push data to your own systems in real time.
Security & Compliance
Is GuardHive PCI DSS compliant?
Yes. GuardHive is PCI DSS Level 1 Service Provider certified. We undergo annual audits and maintain continuous compliance.
How is data encrypted?
In transit: TLS 1.3. At rest: AES-256. Encryption keys are managed by AWS KMS or Azure Key Vault depending on region.
Where is my data stored?
By default, EU data stays in EU (Frankfurt or Ireland) and US data stays in US (Virginia or Oregon). Premium plans include options for specific regions and data residency agreements.
Is GuardHive GDPR compliant?
Yes. We have completed a Data Protection Impact Assessment. Data Processing Agreements are available. We support data deletion workflows and subject access requests.
Do you support PSD2 and 3DS?
Yes. We're PSD2 certified and support Secure Customer Authentication (SCA) step-up triggering. 3DS decisioning is built into the platform.
Can GuardHive integrate with our AML vendor?
Yes. We support API integrations with major AML/KYC providers (Sumsub, IDology, Mitek, and others). Contact us for specific integrations.
Commercial
How much does GuardHive cost?
Pricing depends on volume and feature needs. Freemium is free (800 checks/month), Entry is $499/month (10k checks), Premium is custom. See pricing.
Is there a setup fee?
No. You're live within hours of signing up. No onboarding fees or minimum contracts on Freemium and Entry.
What if I need more than 10,000 fraud checks per month?
Premium is unlimited. Contact sales for a custom quote.
Can I pause my subscription?
Yes, for up to 90 days. For longer pauses, let's discuss your needs.
How do I cancel?
Annual contracts can be cancelled with 30 days' notice. Monthly subscriptions can be cancelled immediately. We'll export your data on request.
Technical Details
What programming languages do you support?
The API is language-agnostic (REST/JSON). We provide SDKs for Python, Node.js, Java, and Go. Any language that speaks HTTP works.
What's your SLA uptime?
Freemium: 99.5%. Entry: 99.9%. Premium: 99.99% (with 24/7 support).
How often do you update models?
Models retrain weekly on the latest transaction data. Updates deploy automatically with zero downtime.
Can I track model performance?
Yes. The dashboard shows precision, recall, false positive rate, and fraud detection rate. The same metrics are exposed via API.
What's your incident response time?
Premium: 1 hour. Entry and below: best effort within 4 hours.
Do you support on-premise or private cloud deployment?
No. GuardHive is a managed SaaS platform. We run it for you, which simplifies operations, ensures uptime, and makes compliance easier.