FAQs
Questions, answered
Everything you need to know about SurveyGuard, from setup and scoring to privacy and pricing.
Getting started
Setting up SurveyGuard
SurveyGuard is an AI-powered, real-time survey fraud detection platform. It checks email reputation, IP signals, device fingerprints and behavior in real time, before a response ever hits your database.
Sign up, create a project for your survey and copy the API key from the project settings. Paste the SDK just before the closing </body> tag of your survey page. Every submission is then analyzed in the background and appears in your dashboard with a Trust Score.
Installing the SDK takes under two minutes. Configuration is optional, because SurveyGuard detects form submissions on its own.
Use the JavaScript SDK, call the REST API from your server, or forward submissions from platforms such as Typeform, Google Forms, Wufoo and WordPress (Contact Form 7). Any survey tool that can send a webhook can be connected. See Integrations for the full list.
Yes. Upload a CSV and clean it after the fact.
Detection & scoring
How responses are scored
Every response is checked across five signal categories: email (disposable domains, spam traps, domain age, MX records), network and IP (VPNs, proxies, datacenter IPs and reputation), device (fingerprinting and duplicate devices), behavior (completion speed, interaction patterns and submission velocity) and answer quality (open-text gibberish and straight-lining).
Each category is scored on its own, then combined into a weighted average from 0 to 100. Higher means more trustworthy, so your team gets one clear verdict per response.
Responses scoring 80 to 100 are low risk and pass straight through. Scores from 50 to 79 are flagged in your dashboard for review. High-risk responses below 50 can be blocked silently when blockHighRisk is on.
Yes. Explain Trust Score gives a plain-language explanation of why a response scored the way it did, the Fraud Pattern Summarizer summarizes patterns across a project, and Anomaly Detection spots unusual spikes before they skew results.
Privacy
Respondents and your data
No. Scores, flags and risk levels are only visible in your dashboard. High-risk respondents get a generic “Unable to process” message, so the analysis can’t be gamed.
Yes. Keep each survey in its own project, with its own API key and stats.
Plans & billing
Pricing questions
No. Sign up for free. You only enter payment details when you choose a plan.
Starter covers up to 5,000 responses a month, Pro up to 25,000 and Unlimited has no cap. See Pricing for the full comparison.
No. Trust Score v2 (email, IP, device and behavioral signals) runs the same on every tier. Pro and Unlimited add the real-time API, open-text gibberish detection and straight-lining detection.
Yes. Upgrade, downgrade or cancel yourself in the Customer Portal. Your service continues through the end of the current billing period.
Still have questions?
Send us a message and the right person on our team will reply by email.