How We Detect Scams
ScamCheckr is a pattern detector that analyzes observable signals to identify scams, phishing, and fraud so you can make informed decisions.
Our Detection Methodology
Every analysis runs through a rigorous detection framework. Our AI engine is trained to identify scam patterns, phishing indicators, and manipulation tactics using evidence-based signals:
Evidence-Based Scoring
Every piece of content starts at a neutral score. We add or subtract based on observable scam indicators — never assumptions.
Pattern Recognition
We compare content against thousands of known scam patterns, phishing templates, and fraud tactics to identify threats.
Observational Language
We use phrases like "consistent with scam patterns" and "indicators detected" rather than definitive judgments. You make the final call.
Multi-Signal Analysis
No single signal triggers a low score. We combine domain reputation, content analysis, manipulation tactics, and cross-references for accuracy.
Transparency Over Certainty
When confidence is limited, we say so. Our Data Confidence badge tells you exactly how much information was available for analysis.
Trust Score Breakdown
The Trust Score is a composite rating from 0 to 100, calculated by weighing the results of each analysis module. A higher score means fewer scam indicators were detected.
Content appears legitimate with strong sourcing, verified domain, and no significant scam indicators detected.
Generally credible but may have minor concerns — unverified claims, light promotional language, or incomplete disclosure.
Notable concerns detected. May include misleading framing, pressure tactics, or suspicious payment requests.
Multiple scam indicators present — phishing patterns, fake urgency, undisclosed promotions, or impersonation tactics.
Content matches known scam patterns and is almost certainly deceptive or fraudulent. Do not engage.
What We Analyze
Every analysis runs through up to 9 independent modules. Each produces its own findings and confidence level, which are combined into the final Trust Score.
Phishing Detection
Checks URLs against known phishing databases, analyzes domain age, SSL certificates, and link patterns to identify fake or malicious websites.
AI-Generated Scam Content
Identifies AI-generated text and images commonly used in scam emails, fake reviews, and fraudulent product listings.
Hidden Promotions & Payment Traps
Detects undisclosed advertisements, fake payment pages, affiliate scams, and deceptive checkout flows designed to steal financial information.
Source Credibility
Evaluates the publishing source — domain age, verification status, editorial reputation, and whether the site appears on known scam registries.
Red Flag Analysis
Flags urgency tactics, too-good-to-be-true offers, impersonation, missing contact info, and other hallmarks of fraudulent content.
Emotional Manipulation
Detects psychological tactics including fake urgency, fear, guilt, false authority, scarcity appeals, and romance scam patterns.
Synthetic Media Detection
Analyzes video and image content for AI-generated or manipulated media commonly used in impersonation scams, fake endorsements, and deepfake fraud.
Account Insights
Profiles the publishing account — age, type, engagement patterns, follower ratios, and behavioral signals indicating bot or scam activity.
News Cross-Reference
Cross-checks claims against trusted news databases and fact-checking organizations to verify legitimacy of offers and claims.
Data Sources & Cross-References
Key claims are cross-checked against established verification databases and domain reputation registries. When a claim matches a known report, we link directly to the source with a reference badge.
Data Confidence Levels
Not all content provides the same amount of analyzable data. We always tell you how much information was available with a transparency badge on every report.
Rich content with text, metadata, domain info, and verifiable claims. Highest analysis accuracy.
Some data available but missing key signals — e.g., no domain info or limited text. Findings may have gaps.
Minimal analyzable content — image-only, very short text, or heavily obfuscated. Treat findings with extra caution.
Got a suspicious link?
Paste any URL, upload a screenshot of an email, or submit suspicious text — your first 3 analyses each day are free.
Start Analyzing