Filter Profane Incoming Messages in P2P Texting Campaigns

Want your agents to keep working a conversation without ever seeing abusive text or explicit images? Here's how to turn on Profanity Filtration for a Peer-to-Peer (P2P) Texting campaign, choose what happens to flagged content, and review it as a moderator.

Replies to texting campaigns can include abusive language or explicit images, and your volunteers are the ones who read them. Profanity Filtration detects profane text and potentially explicit MMS images in incoming messages before they reach your agents. You decide what happens next: hide only the flagged content and let the agent keep the conversation, opt the contact out, send the content to a moderator, or take no action.

The recommended option, Hide flagged content from agents, is built for volunteer-driven teams. Agents are protected from abusive text and explicit images, and you don't have to opt out the contact or reassign the conversation just to shield them. A manager reviews the hidden content later and decides whether it stays hidden. To learn how the AI detection works, see How Profanity Filtration Works and Why It Matters.

Before you start

• Campaign type: Profanity Filtration is available on Peer-to-Peer Texting campaigns only.
• What it detects: profane text and potentially explicit MMS images.
• Access: you need permission to create or edit P2P Texting campaigns. Reviewing flagged content requires a manager account.
• If you choose Route to moderator: decide which agents will act as moderators. You need their usernames during campaign setup.
• Time: about 2 minutes on top of your normal campaign setup.

Choose what happens to flagged content

When Profanity Filtration flags a message or image, it applies the action you pick for the campaign.

Turn on Profanity Filtration

1. While you set up your P2P Texting campaign, turn on Profanity Filtration.
2. Choose the action for flagged content: Hide flagged content from agents, Auto opt-out, Route to moderator, or No action.
3. If you chose Route to moderator, enter the usernames of the agents who will review flagged messages.
4. Complete the rest of your campaign setup as usual.

Pick the action that fits your team. Hide flagged content from agents is recommended.

ⓘ Existing campaigns that already use Profanity Filtration route flagged messages to moderators. Edit the campaign if you want to switch to a different action.

What agents see when content is hidden

With Hide flagged content from agents, the agent keeps the conversation but never receives the flagged content. In its place, the agent sees:

Sensitive content hidden This message was hidden by your organisation's safety settings.

Agents see a placeholder in the original position of the flagged message.

ⓘ Agents can't view or reveal the original content. The conversation stays available to the agent unless a manager reassigns it or opts the contact out.

Review hidden content as a manager

When a conversation has flagged content, it waits in your review queue until you decide what happens to it. The original content isn't shown automatically when you open the conversation.

1. In the Manager Inbox, open the Status filter and select Profanity detected to see conversations with flagged content awaiting your decision.
2. Click the conversation to open it. You see Potentially sensitive content detected and Hidden from agents.
3. Click Review content to reveal the original content to you. You see the detection type and the original text or image.
4. Choose one outcome: Keep hidden, Show to agents, or the existing Opt-out action for the conversation.

Reveal the content, then decide. Revealing does not count as reviewing.

⚠ Choosing Show to agents replaces the placeholder with the original content in its original position, so the agent can then read or see it. Review the content first and confirm you want your agent exposed to it.

ⓘ Revealing content doesn't mark it as reviewed. If you leave without choosing an outcome, the conversation stays in the review queue under Sensitive content review.

Review outcomes

Each outcome removes the conversation from the Profanity detected filter and records your decision. Reviewed content doesn't come back to the filter.

When a conversation has several flagged messages

Each flagged message has its own review status. The conversation stays under Sensitive content review until you resolve every flagged message in it. If you opt the contact out, all pending flagged messages resolve as Reviewed · Contact opted out.

Review flagged messages as a moderator

If you chose Route to moderator, flagged messages follow this path instead.

1. A contact sends a reply containing explicit or vulgar language.
2. The AI identifies the content and blocks it before it reaches the assigned agent.
3. The blocked message goes to the moderator agents' spam folder in the agent console.
4. The moderator checks the message. If it was flagged by mistake, the moderator marks it as non-profane, which releases it to the original receiving agent.

ⓘ With Route to moderator, flagged messages are visible only to the campaign manager (admin console) and the designated moderator agents (agent console), not to the agent the message was headed to.

Find flagged messages and contacts

Use either path to see everything Profanity Filtration has caught in a campaign.

From the Campaign Overview

1. Open your campaign's Overview page.
2. In the Profanity Filter summary, click View messages or See contacts.
3. CallHub takes you to the Responses section with the filter applied.

From the Responses section

1. Go to the campaign's Responses section.
2. Choose the status Profanity detected conversation.
3. Click Apply Filter to see the flagged messages and the contacts who sent them.

Filter Responses to see every flagged message and its sender.

✓ To keep a record, click Export Results with the filter applied. The export includes the flagged messages from the campaign.

Notes & limits

ⓘ Profanity Filtration is available on P2P Texting campaigns only.

ⓘ Customizing filter sensitivity or adding your own custom words isn't available yet.

FAQs

Can agents see the content I hide from them? 
No. Agents see only the Sensitive content hidden placeholder and can't reveal the original content.

Does the agent lose the conversation when content is hidden? 
No. The conversation stays with the agent unless you reassign it or opt the contact out.

What happens to a message that's incorrectly flagged? 
If you chose Route to moderator, the moderator can mark it as non-profane from their spam folder, and it's delivered to the original receiving agent. If you chose Hide flagged content from agents, click Show to agents when you review it.

Who can see the original flagged content? 
Managers, after they click Review content. With Route to moderator, the designated moderator agents can also see it in their spam folder.

Is this available on other campaign types? 
Not yet. Profanity Filtration is supported on P2P Texting campaigns only.