Apps & media

YouTube’s recommendations feel worse in 2026, but these 5 simple tricks fixed mine

At a glance:

  • Use incognito mode or secondary accounts to explore new content without polluting your main feed.
  • Regularly review and update subscriptions to keep recommendations aligned with your interests.
  • Prioritize liking videos over using 'Not interested' to positively influence suggestions.

The problem with YouTube recommendations

YouTube houses hundreds of millions of videos covering a broad range of topics from millions of creators. While this vast content library benefits viewers, it also means YouTube’s recommendations can often miss the mark. In recent weeks, an uptick in disgruntled users online has highlighted this issue, with complaints about videos being re-recommended ad nauseam or suggestions misaligned with their subscriptions and searches. This problem ultimately ruins the YouTube experience, whether users pay for premium or not, as recommendations fail to surface content they’re likely to watch or discover.

The root cause lies in YouTube’s algorithm, which makes assumptions based on watch history, current trends, subscriptions, and searches. Unlike curated streaming services, YouTube’s open ecosystem lacks clear boundaries, leading to inconsistent suggestions. This algorithmic uncertainty creates frustration for viewers seeking a personalized experience, turning YouTube’s strength—its content diversity—into a weakness when recommendations become irrelevant or repetitive.

Keep your main account focused

A curated YouTube feed ultimately depends on user discipline rather than algorithmic perfection. While YouTube’s algorithms play a role, they can only infer preferences from limited signals. To maintain control, I treat my main account as a sanitized space for content I genuinely enjoy. When exploring topics or channels outside my established interests, I use incognito mode or a secondary account. This prevents unrelated content from skewing recommendations on my primary feed.

For example, if I’m curious about a new channel, I’ll watch several videos in incognito mode to evaluate its style and relevance before subscribing. This decision directly impacts future recommendations, so I reserve subscriptions only for channels that align with my core interests. The result is a satisfying feedback loop: focused viewing history attracts similar channels, reinforcing a personalized feed. While I occasionally stray from this rule for highly relevant content, the disciplined approach ensures my main account remains tailored to my preferences.

Leverage YouTube’s search

YouTube’s search functionality is a powerful tool for steering recommendations. By actively searching for topics you enjoy, you signal your interests to the algorithm. For instance, during my balcony gardening phase, searching terms like 'gardening tips' or 'building balcony garden' prompted YouTube to surface more related content. This method works best for core interests, as repeated searches reinforce your preferences in the recommendation system.

However, this approach requires discernment. For fleeting interests—such as catching up on news about the Artemis program—I use a secondary account. This prevents temporary curiosities from permanently altering my main feed. The key is aligning search behavior with long-term content goals, turning YouTube’s search into a precision tool rather than a passive feature.

Use incognito mode, alt accounts, or third-party apps

Separating your digital life into silos minimizes cross-contamination in recommendations. Beyond incognito mode, I maintain a secondary YouTube account for content that doesn’t fit my main interests. This system extends to third-party YouTube apps or browsers designed for anonymous browsing. These tools are essential when:

  • Viewing random videos outside your established interests.
  • Testing channels you’re unsure about.
  • Watching music videos to avoid flooding your feed with music-related suggestions.
  • Opening videos shared by others.
  • Consuming content you don’t want influencing your main account.

This compartmentalization ensures your primary feed remains focused, while secondary accounts handle exploratory or temporary content. The flexibility of this approach adapts to diverse viewing habits without cluttering your core experience.

Manage your subscriptions

Subscribing to a new channel triggers substantial ripple effects in your recommendations. After subscribing, YouTube often floods your feed with content from that channel, making it critical to verify your interest beforehand. To maintain control, I review my subscriptions every few months, dropping channels I no longer watch or enjoy. I also actively 'revive' forgotten subscriptions by watching a video from a dormant channel. This action signals the algorithm to resurface that channel’s content, preventing buried subscriptions from becoming dead weight.

The review process is straightforward: open your subscriptions list, identify inactive or unenjoyed channels, and unsubscribe. For channels you want to keep but rarely see, watching one video reignites their presence in your feed. This proactive management ensures subscriptions remain a strength, not a weakness, in your recommendation ecosystem.

Engage positively with likes

Contrary to popular belief, using 'Not interested' or 'Don’t recommend channel' may not effectively refine recommendations. Instead, I’ve found that positive engagement—particularly liking videos—carries more weight. When you like content, YouTube interprets it as a strong preference, while ignoring videos you dislike signals disinterest without amplifying their presence. This approach benefits both your feed and creators, as likes directly support channels you enjoy.

I reserve 'Not interested' for extreme cases, such as channels with significant AI-generated content or low-quality uploads. Even then, I prioritize the 'Report' option to flag problematic content to Google. For most situations, focusing on likes creates a cleaner feedback loop: your positive interactions guide future suggestions, while negative interactions fade quietly. This strategy aligns YouTube’s algorithm with your actual tastes.

The fresh start: Clear your watch history

If your recommendations feel irreparably skewed, a clean slate is the ultimate solution. Start by clearing your entire watch history. You can further refine this by systematically removing channels from your subscription list, though starting with a fresh watchlist provides immediate relief. This approach is particularly useful if your history contains a mix of conflicting interests or outdated content.

To clear history, navigate to Settings > History & Privacy > Clear watch history. For subscriptions, visit your Subscriptions page and select channels to unsubscribe. While drastic, this method resets the algorithm’s understanding of your preferences. Combine it with the tips above—like using incognito mode for exploration—to rebuild a focused feed from scratch. The result is a YouTube experience that feels curated, not coerced.

Editorial SiliconFeed is an automated feed: facts are checked against sources; copy is normalized and lightly edited for readers.

FAQ

Why are YouTube recommendations worse in 2026?
YouTube’s algorithm struggles with its vast content library, leading to off-mark suggestions. Users report repeated videos, misaligned recommendations, and irrelevant content due to assumptions based on watch history, trends, subscriptions, and searches.
How do incognito mode and secondary accounts improve recommendations?
They prevent exploratory content from polluting your main feed. Use incognito for testing channels or viewing random videos, and secondary accounts for topics outside your core interests. This compartmentalization keeps your primary recommendations focused.
Is liking videos more effective than using 'Not interested'?
Yes. Liking signals strong preferences to the algorithm, while 'Not interested' may not refine suggestions effectively. Save 'Not interested' for extreme cases like AI-generated content, and prioritize likes to positively influence recommendations.

More in the feed

Prepared by the editorial stack from public data and external sources.

Original article