Bored Panda

Instagram 'Great Purge' Cuts Millions of Followers

Instagram has carried out a large-scale cleanup that removed bots, spam and inactive accounts from follower lists across the platform, affecting celebrities, creators and brands. Reports cite very large losses for some high-profile accounts (Bored Panda reports Cristiano Ronaldo lost roughly 6.6 million followers, with some estimates higher). Meta frames the action as part of efforts against "coordinated inauthentic behavior" and Instagram has recently expanded its Original Content Guidelines, reducing visibility for reposted/low-effort content. The removals have sparked reports of legitimate accounts being mistakenly restricted or banned. The story has driven conversations to Meta’s Threads platform and Meta is testing AI-powered support tools (Meta AI Support Assistant) related to moderation. Published by OnlineMarketing.de on 2026-05-08.

MFA Pages: Balancing Ad Revenue and Brand Safety Risks

Made-for-Advertising (MFA) pages are sites designed to maximize ad exposure, defined by the IAB as pages that prioritize ad count and may use manipulative navigation to reach content, often at the expense of the user experience. The article distinguishes MFA from high-quality content, noting that some MFA pages attract real audiences and can be suitable for certain ad placements. It contrasts legitimate and fraudulent MFA practices, citing techniques such as bot traffic, ad stacking, pixel stuffing, and ad slot refreshing that monetize without genuine value. Content on MFA pages is often auto-generated, crawled, or copied from sources like Reddit or RSS feeds, with images frequently reused without permission. Brand safety concerns are raised, as MFA sites may present objectionable material that harms brands. Blocking MFA domains is imperfect due to potential loss of valuable reach, limited blocklists, and new or spoofed domains; targeted analysis is recommended over static blocking.

Keyword Blacklists Hurt Brands: Embrace Contextual Targeting!

An AdExchanger feature, sponsored by Oracle Data Cloud, argues that strict keyword blacklists for brand safety are limited because they lack contextual understanding. The piece notes that many brands rely on IAB categories (arms or weapons, illegal drugs, tobacco, adult content, hate speech, death or injury, and terrorism) but exact-matching keywords can exclude legitimate content. For example, blocking the word 'dead' would also filter references to The Walking Dead, Deadpool, and Grateful Dead, as well as common phrases like 'dead lift' or 'dead end.' The article warns that broad blacklists can divert traffic away from credible publishers toward long-tail sites, hurting premium news brands. It advocates contextually aware targeting that analyzes overall page relevance to balance safety with reach. Quotes from The Wall Street Journal's David Minkin emphasize the need for a strategic, context-driven approach, and the piece concludes that smarter solutions exist to preserve brand safety without sacrificing quality journalism.