X Allegedly Sends Fake Traffic Across the Web
▼ Summary
– X’s new iOS link experiment preloads content before users click, artificially inflating website traffic metrics for platforms like Substack and Bluesky.
– The update collapses posts when links are clicked, allowing continued interaction with X’s engagement buttons while viewing webpages, unlike the previous browser that blocked posts entirely.
– This preloading system distorts analytics by increasing apparent click rates and creating false traffic impressions that may not represent actual human visits.
– X implemented the change to address creator complaints about lower reach for link posts, as the old browser covered posts and reduced engagement signals.
– The metric inflation may harm external creators and publishers by blurring real user engagement and requires platform transparency about how engagement is counted.
A recent technical change on the X platform appears to be generating artificial web traffic, creating confusion for publishers and advertisers who rely on accurate analytics. Websites including Substack and Bluesky reported sudden, unexpected surges in what they describe as “fake” visits following an iOS update from the social media company. According to Nick Eubanks, Vice President of Owned Media at digital marketing firm Semrush, this anomaly stems from a new feature that preloads webpage content before users actively choose to click a link.
Eubanks explains, “This situation represents a clear example of metrics distortion resulting from product experimentation at the platform level.” Under the new system, when a user selects a link within an X post, the post itself collapses rather than disappearing, allowing continued interaction with engagement buttons, such as like, repost, reply, and bookmark, while the external webpage loads. Previously, opening a link in X’s in-app browser would completely obscure the original post, which the company believes reduced engagement with its own content.
“X’s new browser is pre-loading link content in the background, meaning the system fetches the destination page before a human actually taps or views it,” Eubanks notes. He emphasizes that this approach artificially inflates analytics in several important ways. It can misleadingly boost apparent click-through rates and give advertisers, publishers, and creators the false impression of increased human traffic, even when many recorded visits do not correspond to genuine user interest.
Substack CEO Chris Best initially welcomed a noticeable uptick in site traffic after X’s update went live, only to later determine that “most of the apparent lift is fake.” Despite this, Best clarified that Substack still experienced a real increase in visitors after accounting for the artificially inflated numbers.
Bluesky product manager Paul Frazee echoed these concerns, stating that X’s preloading mechanism has “ruined” his platform’s ability to accurately track logged-out daily active users. Frazee added, “X has started to open links in the background to make them load faster… but it has caused a bunch of other sites to get extra traffic that appears real.”
From X’s perspective, the redesign aims to solve a persistent issue for creators. According to X product head Nikita Bier, posts containing links have historically suffered from lower reach because the in-app browser would cover the original post, causing users to forget to engage with it. Bier wrote, “This is because the web browser covers the post and people forget to Like or Reply. So X doesn’t get a clear signal whether the content is any good.”
Although preloading can potentially increase engagement metrics on X itself, it risks harming external creators and publishers by muddying their traffic sources. Eubanks warns, “We’re entering an era where metrics inflation through interface tricks, preloading, autoplay, and AI summarization will blur the line between user engagement and machine behavior.” He suggests that platforms hoping to maintain trust among creators and advertisers must clearly explain how they measure engagement and rigorously distinguish between automated previews and actual human visits.
(Source: The Verge)
