Investigation finds ads on Instagram for several services selling fake followers or automated notification spam for luring in followers
Josh Constine / TechCrunch : Tweets: @counternotions , @joshconstine , @joshconstine , @joshconstine , @joshconstine , and @joshconstine Tweets: Kontra / @counternotions : What do you get when you cross incompetence with immorality? http://techcrunch.com/... Josh Constine / @joshconstine : Social networks fail to reinvest their huge profits into moderation. They're too dependent on tech solutions and don't hire enough humans. It took me just a few minutes to find 5 more spammers selling Insta followers. 5/ http://twitter.com/... Josh Constine / @joshconstine : When popular apps neglect moderation, users pay the price. I found the same problem as I reported on WhatsApp & child sexual abuse imagery 6/ http://techcrunch.com/... Josh Constine / @joshconstine : In response to my investigation, Instagram blocked all 17 services from buying ads & banned some of their accounts. But why did it need to outsource this moderation to a journalist? 4/ http://twitter.com/... Josh Constine / @joshconstine : Instagram supposedly banned spammy follower-buying services & their ads, yet I found 17 of them openly advertising on the app. Some said they paid Instagram $500/day. Unacceptable. 3/ http://twitter.com/... Josh Constine / @joshconstine : Even worse than fake followers on Instagram are apps that charge to take over your account and mass follow/unfollow people to trick them into following you back through notification spam. So I started cataloging them... 2/ http://twitter.com/...
Context & Ripple Effects
Instagram's fake-engagement problem had already moved beyond shadowy third-party tools: earlier reporting documented how real users turned their own accounts into bots through services like Fuelgram that trade likes and comments to game the algorithm. Josh Constine's investigation adds a sharper twist — after Instagram said it had banned follower-selling services, 17 of them were still openly advertising inside the app itself, some reportedly paying about $500 per day for the privilege.
The finding cuts at the platform's core business: Instagram was collecting ad revenue from sellers of exactly the behavior it claims to prohibit, echoing the pattern behind the earlier 125-app Android ad fraud network. Constine frames the root cause as underinvestment in human moderation relative to profits — a critique aimed squarely at how social networks allocate resources.
First-order effects
- Instagram blocked all 17 services from buying ads and banned some of their accounts, directly cutting off a revenue stream some of these sellers had been paying roughly $500 a day to access.
- Brands and creators who buy followers face heightened scrutiny, since paid-audience inflation is now shown to be purchasable through Instagram's own ad system.
Second-order effects
- Rival platforms get pulled into the same cleanup: within weeks of this investigation, Twitter suspended three notification-spammer apps for violating its API rules on aggressive following and follow churn — the same mass follow/unfollow tactic the Instagram ads were selling.
- Enforcement-by-headline becomes the operating model: as with scammers who later abused Instagram's impersonation protections to charge $60 per targeted account ban, action follows press exposure rather than proactive detection.
Third-order effects
- If moderation budgets stay dependent on tech solutions instead of human review, platforms will keep monetizing spam economies until each investigation forces a reactive purge — making trust in engagement metrics, not just content, the infrastructure they must defend.
- The persistent market for synthetic followers points toward a structural split between platforms whose ad systems police their own inventory and those whose revenue depends on looking away.
The trend: Social platforms' ad businesses keep profiting from the same fake-engagement economies they claim to ban, with enforcement arriving only after investigative pressure forces it.