Research by a group of Italian security analysts shows that the photo-sharing network still has a significant spam and fraud problem.
There’s no question that Instagram has become a massive platform for sharing images since it was acquired by Facebook in 2012 for $1 billion: According to the company, it has over 300 million active users, and more than 70 million photos are shared every day. Over the past year, Instagram has made a number of moves to turn that user base into a revenue-generating machine and a full-fledged media platform, by adding advertising and news-related features like its recently launched “Explore” page.
In the midst of all this growth, however, the company has also been fighting a rising tide of spam and fraud. It recently purged a huge number of fake accounts, and many users saw their follower count plummet at a result. But despite that purge, Instagram still has a significant problem with fake and spam accounts, according to a research paper published earlier this month.
The researchers, a group of Italian security analysts, looked at more than 10 million Instagram users who had posted a total of over 570 million images, likes, and comments. They also bought about 20,000 spam-bots or automated accounts from 10 different online vendors. Based on this research, their conclusion was that almost 30% of the service’s accounts are inactive (one item posted or less), and an estimated 8% of the total behave like automated spam or fraudulent accounts.
In their paper, the researchers describe the huge shadow ecosystem that has grown up around Instagram, with brokers offering to sell thousands or even hundreds of thousands of fake accounts, as well as fake or spammy activity such as likes and follows. One ad offers “100% real people, no password required, money-back guarantee” and promises delivery of 1,000 followers for $12.99. Another service offers 2,000 followers on the platform for $19.99 and promises delivery in 1 to 5 days.
In total, the researchers said they found close to 200 services that were selling followers as well as activity. Many are just websites with spammy marketing claims, but others are fairly sophisticated operations that use freelance-work or crowdsourcing platforms such as Fiverr and SeoClerks to provide content and users for their operations, and offer payment options like PayPal and Bitcoin. Their servers are set up to add followers, likes, and comments from thousands of fake accounts.
Many of these providers have apparently also developed sophisticated software and hardware solutions to the restrictions that Instagram imposed in its recent purge. They offer the equivalent of spam bot-nets, with multiple IP addresses, geographically separate and even automatically randomized.
As the paper points out, the kind of social following that prominent users can command on platforms like Instagram, Facebook FB , and Twitter TWTR can translate directly into advertising and promotional revenue, since brands of all kinds are increasingly interested in users who have large followings, as a way to generate word-of-mouth marketing for their products. Some hotels, for example, will offer free rooms and other services to users who have a large number of followers or activity on their accounts.
Major brands are already spending millions of dollars a month on Instagram because their content appears to generate much higher levels of engagement—clicks, likes and comments—from users on that platform compared to Twitter or even Facebook. And the company continues to appeal to advertisers with new features and offerings. Based on the volume of users and revenue the company is pulling in, some analysts say Instagram’s market value could be as high as $35 billion. But how much of that is based on fraudulent or spammy activity?
The Italian research report seems to indicate that this kind of behavior still makes up a significant proportion of Instagram’s user base, even after its attempt to clean house. And that could be a problem for the company as it tries to expand its ad and media business.