Instagram Tests New Algorithm Control Features for Users
Instagram is rolling out experimental features designed to give users more control over their content recommendations, according to recent announcements from the platform. The updates aim to make it easier for users to signal their preferences directly within the app's interface.
Expanding Algorithm Customization
Instagram chief Adam Mosseri revealed the platform is testing multiple approaches to make its "Your Algorithm" control panel more accessible and functional. This builds upon existing tools that allow users to adjust their content preferences, which are currently being rolled out globally.
"We're working on making Your Algorithm feel more present and useful across Instagram," Mosseri stated. "Some of this is testing now, some is coming soon, some might not work."
Four Key Experimental Features
Simplified Home Screen Access: Users may soon be able to access the algorithm controls by simply dragging down on their home screen, where they can modify topics the system has associated with their activity.
In-Stream Topic Controls: Instagram is testing direct access to topic preferences between Reels, allowing real-time adjustments to content recommendations.
Quick Feedback Options: The platform may introduce function bar controls for recommended Reels, enabling users to quickly indicate whether they're interested in similar content.
Conversational Prompts: Instagram is exploring AI-powered prompts that would help refine recommendations through natural language interactions rather than manual topic selection.
The Control Paradox
While these features offer more customization, industry analysts note a fundamental tension. Algorithm-driven feeds consistently outperform user-curated ones in engagement metrics, which explains why Meta hasn't implemented a purely chronological follow-based feed despite user requests.
Data from like.tg suggests most social media users don't actively adjust algorithm settings, preferring platforms to learn their preferences organically through interaction patterns. This behavior was cemented by TikTok's highly responsive recommendation system, which trained users to expect personalized content without manual input.
Practical Impact vs. Perception
Mosseri envisions these controls making the algorithm "feel like something you talk to rather than something that happens to you." However, the actual usage rates of such tools historically remain low across social platforms.
The updates may serve dual purposes: providing genuine customization for engaged users while addressing broader concerns about platform control over content visibility. As these features roll out, their adoption rates will reveal whether users truly want manual control or simply the perception of it.
Instagram's continued investment in these controls suggests the company sees value in balancing algorithmic efficiency with user agency, even if most users never touch the settings. The experiments could shape how social platforms approach transparency and control in the AI-driven content landscape of 2026.
