Comment sections have been declared dead dozens of times over the past decade. Many major publications disabled comments entirely, citing toxicity and moderation costs. Others maintain comment sections that are barely moderated wastelands of spam and hostility. A few sites have comment sections that actually work—thoughtful discussions that add value rather than detract from content. The difference isn’t luck. It’s moderation strategy.

Why Unmoderated Comments Fail

The pattern is predictable. A site launches with open comments. Early commenters are often thoughtful contributors from the site’s core audience. Discussion quality is good.

As the site grows, comment volume increases. Toxic commenters appear—trolls seeking reactions, aggressive debaters, spammers. Without moderation, these users drive away thoughtful commenters who don’t want to engage with hostility.

The comment section enters a death spiral. Good commenters leave. Bad commenters dominate. New visitors see toxic comments and either join the toxicity or avoid commenting entirely. Within months or years, the comment section becomes useless or actively harmful to the site’s brand.

This isn’t inevitable—it’s the result of inadequate moderation. But moderation costs money and effort, which is why many sites either disable comments or let them degrade.

The Three Moderation Models That Work

Heavy pre-moderation: All comments are reviewed and approved before appearing. This is labor-intensive but ensures comment quality. The Wirecutter uses this approach—comments appear hours or days after submission, but quality is consistently high because everything is screened.

Community moderation with clear guidelines: Users flag problematic comments, and moderators review flagged content. Clear, enforced guidelines specify what’s acceptable. Reddit’s better-moderated subreddits use this model successfully.

Technical barriers combined with light moderation: Require user accounts, implement rate limiting, use spam filters, and have moderators remove the worst content without reviewing everything. Hacker News uses this approach—technical barriers reduce low-effort toxicity while light moderation catches edge cases.

All three models require resources and commitment. What doesn’t work is hoping that open, unmoderated comments will magically stay civil through goodwill alone.

Why Most Moderation Fails

Sites that fail at moderation usually make predictable mistakes:

Unclear or unenforced guidelines: Having comment policies that aren’t consistently enforced is worse than having no policies. Inconsistent enforcement creates perception of bias and gives toxic users ammunition to claim unfair treatment.

Insufficient moderator capacity: One moderator for 10,000 comments daily doesn’t work. Sites either need to limit comment volume (pre-moderation, rate limiting) or invest in sufficient moderation capacity for the volume they permit.

Reactive-only moderation: Only removing comments after they’re reported means bad content sits visible for hours or days, damaging discussion quality and driving away good commenters. Proactive moderation—reviewing new comments before they spiral—prevents problems rather than cleaning up afterwards.

Treating all content equally: Not all articles need comments. Highly contentious topics (politics, controversial social issues) require more moderation resources than straightforward articles. Sites that enable comments uniformly across all content spread moderation resources too thin.

The Registration Barrier

Requiring registration before commenting dramatically reduces spam and low-effort toxicity. Anonymous commenting attracts drive-by hostility. Registration doesn’t eliminate bad behavior but raises the friction enough to deter casual trolls.

The tradeoff is reducing comment volume overall. Some potential commenters won’t create accounts just to leave a single comment. For sites that value comment quality over quantity, this is the right tradeoff. For sites using comment volume as an engagement metric, it’s problematic.

Sites can split the difference—allowing read-only viewing of comments without registration but requiring registration to post. This preserves public accessibility while raising the barrier for participation.

The Platform Question

Sites debate whether to use native commenting systems or platforms like Disqus. Disqus provides better moderation tools and spam filtering out-of-the-box. But it creates cross-site commenting networks where toxic users banned from one site can comment elsewhere, and it places ads in comment sections unless you pay for premium tiers.

Native commenting gives you full control and keeps user data on your platform. But you’re responsible for building spam filtering, moderation interfaces, and dealing with regulatory requirements (GDPR compliance, handling takedown requests).

For small sites, platforms like Disqus or Facebook Comments provide moderation infrastructure you couldn’t build yourself. For larger sites with resources, native commenting allows better integration with site design and audience data.

The Community Effect

Comment sections benefit from having a core group of regular, thoughtful commenters. These regulars set tone, welcome newcomers, and self-police bad behavior through social pressure.

Building this community requires consistent moderation that makes clear what behavior is valued. Early commenters become models for later arrivals. If early commenters are thoughtful and civil, newcomers tend to match that tone. If early commenters are aggressive and hostile, that becomes the established culture.

Sites that successfully build comment communities often have moderators who participate in discussions, not just enforce rules. This visible presence signals that comment sections are valued and monitored, which encourages better behavior.

When to Disable Comments Entirely

Some content types don’t benefit from comments. Breaking news articles about tragedies attract performative outrage and inappropriate comments. Deeply personal essays get invasive or judgmental responses. Sensitive topics that should be handled carefully become flame wars.

Disabling comments on specific articles or categories isn’t admitting failure—it’s recognizing that not all content needs or benefits from public discussion. The New York Times disables comments on articles likely to attract harassment or where comments wouldn’t add value, while keeping them enabled on analysis pieces and features where discussion enhances content.

The AI Moderation Question

AI content moderation tools can flag spam, hate speech, and guideline violations. They’re fast and cheap compared to human moderators. But they struggle with context, sarcasm, and subtle violations. They also make mistakes—flagging legitimate comments and missing actual problems.

The effective approach is using AI as first-pass filtering, with human moderators reviewing flagged content and randomly sampling non-flagged comments to catch AI errors. AI handles scale; humans handle judgment.

Relying entirely on AI moderation creates problems when false positives frustrate legitimate users and false negatives allow toxic content to slip through. The technology isn’t yet good enough to replace human judgment, but it’s valuable for managing volume.

Why Comments Still Matter

Despite the “death of comments” narrative, comment sections serve valuable functions:

• Community building: Regular commenters form relationships and ongoing discussions that keep them returning to the site.
• Reader feedback: Comments provide immediate reaction to content quality and identify errors or missing context.
• SEO value: Comment content adds unique, regularly updated text that search engines value.
• Engagement metrics: Comments signal that readers care enough to engage beyond passive reading.

These benefits only materialize with adequate moderation. Toxic comment sections provide negative value—they harm brand reputation, drive away readers, and create legal and reputational risks.

What Working Comment Sections Look Like

Well-moderated comment sections share characteristics:

• Clear, enforced guidelines consistently applied
• Visible moderator presence
• Registration requirements that reduce drive-by toxicity
• Technical barriers (rate limiting, spam filtering) that reduce volume to manageable levels
• Regular participants who model good discussion behavior
• Selective comment enabling—not every article needs comments

These don’t happen by accident. They require resources, commitment, and treating comment sections as integral to site quality rather than afterthoughts.

Comments aren’t dead, but the assumption that you can enable comments and let community goodwill maintain quality is. Sites that invest in proper moderation can have valuable comment communities. Sites that won’t invest in moderation should disable comments rather than letting them degrade into toxic wastelands that harm more than they help.