Adult content accounts for 60 percent of blocked ads, report

PropellerAds has published its Ads Safety Report 2025, detailing how moderation, enforcement and fraud patterns developed across its platform during the year.

The report outlines preventive controls applied before campaigns go live, confirmed account suspensions, and the main fraud tactics identified by its Policy and Security teams.

The company recorded a 35 percent year-over-year increase in campaigns flagged and declined during moderation in 2025. According to the report, this rise reflects expanded preventive controls and broader moderation coverage rather than a surge in unique violating campaigns.

Preventive moderation trends in 2025

The figures represent the number of rejections or blocks, not unique rejected or blocked campaigns. A single campaign may therefore account for multiple rejections.

Adult, pornographic, sexual, or erotic content accounted for 60 percent of all rejects or blocks, totaling 439,927 cases. Antivirus alerts or malware detected on campaign-related domains represented 26 percent, or 191,103 rejections. Automatic file downloads on desktop or mobile devices made up 3.4 percent (24,500), while other policy violations accounted for 3.0 percent (22,127).

Other categories included region-specific restricted content (2.1 percent), trademark violations (1.7 percent), inaccessible destination URLs during moderation (1.1 percent), prohibited products (1.0 percent), malware-related scare claims (0.7 percent), and deceptive financial advice promises (0.5 percent).

Confirmed enforcement and account suspensions

Account suspensions apply to confirmed, high-risk, or repeated violations identified through automated systems and expert review. In 2025, cloaking represented 78.2 percent of all suspensions, with 1,311 accounts affected.

Malware-related violations accounted for 7.6 percent (127), followed by other violations at 6.0 percent (100). Additional suspension reasons included ransomware attacks (2.9 percent, 49 cases), failed KYC due to fake ID submissions (2.6 percent, 44), scam landing pages (2.0 percent, 33), and confirmed fraud involving multiple signals (0.7 percent, 12).

Fraud tactics shift toward infrastructure abuse

The report identifies four major fraud patterns in 2025: infrastructure-heavy cloaking using multi-layer routing and conditional delivery; malvertising through direct file distribution; hijacking attempts targeting Telegram and WhatsApp messenger accounts; and the use of compromised infrastructure, such as expired domains or breached servers.

From a technical perspective, around 80 percent of identified attack vectors targeted Windows and Android users.

PropellerAds states that its internally developed systems rely on AI and machine learning to support anomaly detection, analyze interaction patterns, and prioritize high-risk signals. Infrastructure checks and expert review remain central to enforcement decisions.

www.propellerads.com

Follow Marketing Report on LinkedIn