# Deconstructing social media impersonation scams: A mixed-methods investigation on various strategies used by attackers

> A mixed-methods investigation into the strategies used by attackers in social media impersonation scams.

[Paper page](https://vsl-landing-page-el7cn5p1w-ifly-leonards-projects.vercel.app/research/social-media-impersonation-scams)

## Metadata

- Authors: Leonard Selvaraja Fernando
- Year: 2024
- Organization: Independent research
- Type: Investigative Research Paper
- Category: Research Paper
- Status: Actively researching

## Abstract

Social media impersonation scams have become increasingly sophisticated, exploiting both technical vulnerabilities and human cognitive biases. This paper presents a mixed-methods investigation that combines quantitative analysis of 1,200+ reported scam incidents with qualitative interviews of 40 victims and 12 security researchers. We identify seven distinct attack strategies, ranging from deepfake-assisted identity theft to context-aware phishing campaigns that leverage publicly available personal data. Our findings reveal that attackers increasingly employ multi-stage approaches, combining reconnaissance, trust-building, and exploitation phases. We propose a comprehensive taxonomy of impersonation techniques and evaluate the effectiveness of current countermeasures across major platforms. The study concludes with actionable recommendations for platform designers, policymakers, and end-users.

## Proposed Hypotheses

- H0: There is no significant relationship between attack strategy complexity and victim susceptibility in social media impersonation scams.
- H1: Multi-stage impersonation attacks that combine reconnaissance with trust-building tactics result in higher victim engagement rates than single-stage attacks.
- H2: Victims with higher digital literacy scores are not significantly less susceptible to context-aware phishing campaigns that leverage publicly available personal data.
- H3: Platform-specific countermeasures show differential effectiveness across attack strategies, with visual verification tools outperforming text-based flagging systems for deepfake-assisted attacks.

## Table of Contents

1. Introduction
2. Literature Review
3. Methodology
4. Attack Strategy Taxonomy
5. Quantitative Findings
6. Qualitative Insights
7. Discussion & Implications
8. Conclusion & Recommendations
9. References

## Data Collection Method

Mixed Methods: Quantitative analysis of 1,200+ reported scam incidents from cybersecurity databases and platform trust & safety reports, combined with semi-structured qualitative interviews with 40 victims and 12 security researchers across six countries.

## Tools Used

Not specified.

## Meta

Canonical LLMs file: https://vsl-landing-page-el7cn5p1w-ifly-leonards-projects.vercel.app/research/social-media-impersonation-scams/llms.txt