Skip to content

Free tool · No signup

Write a comment moderation policy your team can follow

A policy is not a list of banned words. It says who decides, how fast, what happens when a thread turns hostile, and how a reader appeals. Answer four questions and get a draft you can publish, copy, or download.

Short answer

A good comment moderation policy says who may moderate, what is published, how held comments are reviewed, how quickly, what happens when a thread turns hostile, and how a reader appeals a decision. It should be short enough to follow during a shift and specific enough to settle the same case the same way twice.

The generator

Answer four questions. Get a policy you can publish.

Nothing is sent anywhere. Your answers stay in this browser tab, and the draft updates as you choose.

Start here

The basics

Question 1 of 4

What kind of site is it?

Question 2 of 4

Who moderates?

Question 3 of 4

When does a comment publish?

Question 4 of 4

What happens when a thread turns hostile?

Approve from email or Slack

Adds a clause that says decisions can be made away from the dashboard and are still recorded.

This is a working draft, not legal advice. Have someone who knows your jurisdiction review it before you publish.

See your policy

Live preview

News or niche media site · Version 1.0

Comment policy

this site

A written comment policy for this site: who moderates, what is published, how fast held comments are reviewed, and what happens when a thread turns hostile.

Who moderates

  • Moderation lead. Sets this policy, owns the queue's response times, and reports on moderation every month for this site.
  • Moderators. Approve, reject, mark spam, and escalate. Two moderators cover a busy thread so no single person decides alone.

Purpose and scope

  • This policy explains how this site reviews, publishes, and removes comments. It applies to every comment box on the site, including articles, guides, and reference pages.
  • We host comments to make the work more useful to readers, not to maximize engagement metrics. Criticism of this site is allowed. Abuse of other readers is not.
  • This is a working document for moderators and a public statement for readers. It is not legal advice, and it does not replace the law or a platform’s terms.

What we publish

  • We publish first-hand experience, questions, corrections, and disagreement with the article. A comment does not need to agree with us to stay up.
  • We publish relevant links when they add evidence or context to the thread. We remove undisclosed promotion, affiliate links, and the same comment repeated across threads.
  • We do not publish harassment, threats, hate speech, targeted abuse, doxxing, impersonation, spam, deliberate disruption, or content that breaks the law or a court order.
  • We remove a comment for breaking one of those rules, not because it is critical. When a comment is removed, the moderator records which rule applied.

How comments are reviewed

  • A first-time commenter's comment is held for review before it publishes. Once someone has an approved comment and no recent removals, later comments publish immediately and are screened by the spam filter.
  • Automated screening runs before the queue. It flags likely spam and holds it; it does not make the final decision on a borderline comment.
  • Moderators may approve or reject from email or Slack. The decision is recorded in the dashboard with the moderator’s name and the time.
  • No moderator acts on a thread they are personally involved in. It goes to another moderator.
When a thread turns hostile

Start with the lightest step that settles the thread. A hold on one thread is reversible; a ban is not, so it comes later.

  1. Hold new comments for review on that thread only. Existing comments stay visible and the rest of the site is unaffected.
  2. Remove the comment that broke the policy and record which rule it broke.
  3. Lock the thread if the discussion cannot continue without abuse. A lock stops new comments; it does not delete the discussion.
  4. Ban the commenter from commenting on this site. A ban applies to one site, never to every site on the platform.
  5. Record the decision — the comment, the rule, the moderator, and the time — and tell the moderation lead.

Response times

  • We aim to review held comments within four business hours during coverage. Reports of threats, doxxing, or illegal content are handled within one hour when a moderator is on shift.
  • These are targets, not guarantees. If the queue is longer than usual, we say so in the thread rather than leave readers guessing.

Bans and appeals

  • A ban applies to one site, not to every site on the platform.
  • A reader can appeal a decision by contacting the moderation contact published on the site. Include the comment and the thread.
  • One person who did not make the original decision reviews the appeal.
  • If we got it wrong, we restore the comment and say so in the thread.

Privacy and data

  • We collect the minimum needed to publish and moderate a comment: display name, comment text, timestamp, and the sign-in identity the reader chose.
  • We do not sell comment data or use it for advertising.
  • We keep moderation records for 12 months, then delete them unless a legal hold applies.
  • A reader can ask for a copy or deletion of their comment data by contacting the moderation contact published on the site.

Reviewing this policy

  • The moderation lead reviews this policy every six months and after any serious incident.
  • Changes are published on this page with the date. Moderators are briefed before a change takes effect.
  • The moderation log is reviewed monthly for patterns: which rules fire, which threads need a second moderator, and whether the response times are realistic.

Contact

the moderation contact published on the site

Questions, reports, and appeals. For urgent safety issues, use the subject line URGENT.

Draft generated with the free EchoThread comment moderation policy generator. Edit it to match your site, publish it, and review it every six months.

The preview and the downloaded Markdown contain the same clauses.

What the draft covers

The four jobs a comment policy actually has to settle

Most published comment policies are a list of prohibited topics. That list does not tell a moderator what to do at 9pm on a Friday. The generator writes the parts that answer the shift-level questions.

Who decides, and who covers

Named roles rather than “the team”: who works the queue, who can ban, who signs off on a removal, and what happens when the usual moderator is away.

Review model and response times

Whether a first-time commenter is held, every comment is held, or comments publish after screening — plus the response time readers can expect and moderators are measured against.

The hostile-thread ladder

The order of holds, removals, locks and bans, so the same behaviour gets the same response from different moderators on different days.

Bans, appeals, and privacy

A ban applies to one site, an appeal is reviewed by someone who did not make the original call, and the policy says what comment data is kept and for how long.

How to use it

From draft to a policy people follow

  1. 1

    Answer for the site you have, not the one you plan to have

    A two-person blog that promises two-hour review times will miss them. Pick the review model and response time you can actually cover this month; tighten them when the team grows.

  2. 2

    Publish it on a URL you control

    A comment policy needs a stable home — /comment-policy, /community-guidelines, or a section on your About page — and a link from the comment box. A policy buried in a PDF is not a policy.

  3. 3

    Put the escalation ladder in onboarding

    New moderators should read the ladder before their first shift and use it on a sample thread. Review the policy every six months, and after any incident that needed a second decision-maker.

The product, honestly

The policy says what should happen. EchoThread gives moderators a place to do it.

The tool is platform-neutral: keep the policy whether you use EchoThread, WordPress, Ghost, or a custom build. If you do use EchoThread, these are the clauses the product already enforces.

  • Every comment is screened at submit time by an integrated ML classifier (Siftfy), after the site owner’s restricted-words list. Borderline comments go to one shared queue.
  • Starter and above add Approve / Reject / Spam links in email, plus Slack or Discord alerts, so a decision can be made away from the dashboard.
  • Thread lock, per-site commenter bans, trusted commenters, and auto-close are available on every plan, including the free Hobby tier.
  • Pro adds a restricted Moderator seat that can review comments without seeing billing or site settings.

Questions

Before you publish it

What should a comment moderation policy include?
Who may moderate, what is published, how held comments are reviewed, how quickly, what happens when a thread turns hostile, how bans and appeals work, and how the policy itself is reviewed. Keep it short enough to follow during a shift and specific enough that two moderators reach the same decision on the same comment.
How long should a comment moderation policy be?
One page is enough for most sites. The generator produces roughly 700–900 words; cut any clause your team will not use. A short policy that moderators actually read beats a long one nobody opens.
Should we pre-moderate every comment?
Only if you have the cover to clear the queue within the response time you publish. Most sites start by holding a first-time commenter’s comment, then let approved commenters publish immediately while automated screening handles spam.
How do we handle a hostile comment thread?
Use the lightest step that settles it: hold new comments on that thread, remove the comment that broke the policy, lock the thread, then ban the commenter. Record the rule and the decision. A second moderator should review a ban longer than 30 days.
Is this policy legal advice?
No. It is a drafting tool. Have someone who knows your jurisdiction review the result, especially for defamation, privacy, and data-retention rules. The generator also does not create a contract with readers.
Does it work for WordPress, Ghost, or a custom site?
Yes. The policy is platform-neutral: it describes the moderation job, not the comment system. Keep the same clauses if you change tools, and change only the parts that name how approvals or bans are carried out.