Today’s audiences spend time endlessly scrolling social media, but the trusted media experience must be different. Readers turn to established news organizations because they trust the source. They expect responsible reporting, editorial insight, and a focus on public interest — all of which is lost when a click-chasing algorithmic approach is employed.
The Reuters report, “Journalism, Media, and Technology Trends and Predictions 2025,” highlights how informed personalization and audience engagement are essential for building loyalty and driving long-term success. Publishers who embrace these strategies and use their data effectively are better positioned to meet audience needs, grow first-party data, and boost subscription rates.
Recently, Facebook reduced its commitment to fact-checking, and also refined its feeds to show users more of the same political content they already consume.
This trend underlines the urgent need to stand apart from social platforms. If news sites adopt the same type of recommender algorithm, they risk losing their editorial identity.
Informed personalization provides a path forward. By embedding journalistic values and editorial oversight into the recommender system, newsrooms can maintain the trust that sets them apart from social media. Below, we’ll examine the pitfalls of existing recommender algorithms and show how informed personalization solves them.
Two major pitfalls of personalization today
Black box algorithms
Conventional recommender algorithms are effectively “black boxes.” They learn from user clicks and engagement without revealing how they prioritize content. This lack of transparency poses two serious threats:
- Eroded editorial control: Journalists lose influence over which topics and stories are highlighted if algorithms simply chase the most popular topics.
- Reduced reader trust: If people feel the site is driven by the same click-based priorities as a social feed, they question its credibility.
When editorial media relies on a purely automated recommender algorithm, it begins to resemble social media. It loses the distinct editorial voice that readers expect from a trustworthy news source.
Echo chambers
Another hazard is the creation of echo chambers. Algorithms that serve “more of the same” can trap readers in content silos, preventing them from discovering essential stories outside their usual interests. For media organizations — which prides itself on a broad, informed perspective — this is a direct clash with its mission:
- Narrowed coverage: Readers may see content that matches their past clicks rather than content editors deem truly newsworthy.
- Weakened public discourse: Journalism aims to inform and challenge assumptions, not just reinforce them.
Both pitfalls blur the line between editorial media and social media. If a news site’s personalization is no different from a social feed’s, why would readers continue to trust it?
Enter informed personalization
Responsible media executives need to push back against the click-driven personalization that social media has normalized. Our purpose is not wasting readers’ time with endless scrolling — it’s about using it to make them smarter, more informed (and yes, sometimes entertain or divert them). This is why the algorithm must be informed by editors so it starts working for us, not the other way around.
This is why I believe it’s time to introduce a new term:
Informed Personalization is a transparent, ethical approach to content personalization in news media, where algorithms are built in close collaboration with editorial teams to maintain journalistic integrity, foster diverse perspectives, and avoid echo chambers. It discards the “black box,” offering an open, controlled system that strengthens audience trust.
Let me give some examples where I believe journalists play a crucial role in informing the recommender system:
When engagement doesn’t equal importance
The most important article isn’t always the one that performs best in terms of clicks. I’ve often seen stories of critical importance chosen by editors struggle to compete with lighter, more trivial pieces in terms of engagement. That’s why it’s crucial for editors to have the ability to override algorithmic recommendations, just as naturally as they prioritize stories on a homepage.
Example: It should be possible to pin articles on top of any recommendation
Keeping evergreen stories alive
Older articles can still hold great value, especially if they’re unique to your brand. The challenge is ensuring that these pieces don’t get buried by fleeting, short-lived news with limited shelf life. These evergreen stories are particularly important for infrequent or first-time visitors, who may not have encountered them before – making them the perfect candidates for a well-designed recommender system.
Creating variation
Articles can be tagged with attributes like section, tone, user needs, or content type—many of which can be automated by AI. These tags are key to keeping recommendations varied and readers engaged. People lose interest quickly, so it’s important to offer as much variety as possible, not just in content but also in how it’s presented.
Pro tip: If your recommender system tends to suggest multiple articles from the same section, try mixing in the best articles from other categories. Readers often prefer fresh, breaking stories from other areas over older content in the same section.
The best systems also play with design to create variation, using different layouts and formats to keep things fresh. A big trend we’re seeing now is how publishers even create multiple versions of the same content, tailoring it for different users to make it feel more relevant.
Conclusion
At a time when social platforms reduce fact-checking and feed users more of the same political content, informed personalization stands as a beacon of responsible, human-driven curation. It respects editorial judgment, fights echo chambers, and aligns with broader goals like loyalty and subscription growth. Above all, it keeps journalists in charge of the news experience. Remember, you’re in the media business. But your currency is trust.
