> ## Documentation Index
> Fetch the complete documentation index at: https://docs.exerta.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Moderation Rules: Control What Gets Hidden or Flagged

> Build no-code moderation rules to automatically hide, flag, or escalate comments by content, author, or ad placement — then track every match.

Moderation rules let you define exactly what happens to each type of comment — without writing a single line of code. You describe the comment you want to catch, choose what should happen to it, and Exerta handles the rest. Each rule is made up of conditions and an action, runs in priority order, and starts working the moment you save it.

## Rule anatomy

Every moderation rule has three parts:

1. **Conditions** — what triggers the rule. You can match on:
   * **Content** — keywords or phrases in the comment text (e.g., "buy followers", a competitor brand name, or a URL)
   * **Author** — who wrote it (e.g., first-time commenter, repeat poster, or a specific account)
   * **Placement** — where it appeared (e.g., a specific campaign, ad set, page, or organic post)

2. **Action** — what happens when all conditions are met:
   * **Hide** — the comment is hidden from public view instantly
   * **Flag for review** — the comment is surfaced in your moderation queue for a human to review
   * **Escalate to DM** — the agent sends the commenter a private message and alerts your support team
   * **Custom Reply** — the agent posts a specific reply you've written for this rule

3. **Priority** — rules run in the order you set. The first matching rule wins, so put your most specific or most critical rules at the top.

## Creating a rule

<Steps>
  <Step title="Go to Moderation → Rules">
    In your Exerta dashboard, navigate to the **Moderation** section and open the **Rules** tab.
  </Step>

  <Step title="Click New Rule">
    Hit **New Rule** to open the rule builder.
  </Step>

  <Step title="Set your conditions">
    Add one or more conditions. For example: comment contains "spam", "link in bio", or a competitor brand name. You can also filter by author type or ad placement.
  </Step>

  <Step title="Choose an action">
    Select what happens when the rule matches: **Hide**, **Flag**, **Escalate**, or a **Custom Reply** with your own message text.
  </Step>

  <Step title="Set priority and save">
    Drag the rule to the correct position in your priority list, then click **Save**.
  </Step>

  <Step title="The rule is active immediately">
    Your rule goes live right away. Every new comment on your connected placements is checked against it in real time.
  </Step>
</Steps>

## Combining conditions

You can stack multiple conditions inside a single rule using AND logic — all conditions must be true for the rule to fire. This keeps your rules precise enough that you never accidentally hide a real customer's comment.

For example, you might hide a comment only if it **contains a link** AND is **from an account with fewer than 10 followers**. Or flag a comment only if it **mentions a competitor** AND **comes from a first-time commenter** on a specific campaign. The more conditions you add, the more targeted — and safe — the rule becomes.

## Per-rule analytics

Every rule tracks its own performance so you know exactly what it's catching:

| Metric               | What it shows                                    |
| -------------------- | ------------------------------------------------ |
| **Comments matched** | Total comments the rule checked and triggered on |
| **Hidden**           | Comments removed from public view by this rule   |
| **Flagged**          | Comments sent to your review queue               |
| **Escalated**        | Comments routed to DM or handed off to your team |

Use this data to spot over-triggering rules (high match count, few real issues) and tighten their conditions, or to find under-performing rules that might need broader keywords.

## Pausing and testing

You can pause any rule with a single click — it stops running immediately, and none of the other rules are affected. To resume, click the toggle again.

Before activating a new rule, use **Preview mode** to test it against your most recent comments. Preview shows you which real comments would have been caught, so you can fine-tune conditions before anything goes live.

<Tip>
  Start with a spam rule that catches the most common patterns ("buy followers", "link in bio", competitor URLs). You can always add more specific rules later once you see what's actually hitting your ads.
</Tip>

<Note>
  The Starter plan includes **7 moderation rules**. Scale and Enterprise plans support higher rule limits. Comment hiding rules (with stock replies) are available on the Free plan.
</Note>


## Related topics

- [What Is Exerta? AI Engagement Agents for Social Ads](/introduction.md)
- [Get Started with Exerta: Live AI Agents in 5 Minutes](/quickstart.md)
- [Multi-Branch Logic: Route Comments by Intent and Sentiment](/moderation/multi-branch-logic.md)
- [Scope Moderation Rules to Campaigns, Ad Sets, and Ads](/moderation/campaign-targeting.md)
- [Comment Moderation: Protect Your Ads Automatically](/moderation/overview.md)