Documentation

Message Moderation

TipPanel provides powerful moderation tools to help you manage user messages, filter inappropriate content, and maintain a positive environment for your audience.

Table of Contents
  1. Moderation Overview
  2. Moderation Queue
  3. Banned Words Filter
  4. AI Moderation
  5. Moderation Settings
  6. Moderation Tips and Best Practices

Moderation Overview

TipPanel offers three levels of message moderation to help you manage user-generated content:

Moderation Queue

A dashboard where you can manually review, approve, or reject messages before they appear publicly.

Banned Words Filter

An automated filter that blocks messages containing specific words or phrases you define.

AI Moderation

Advanced AI-powered content moderation that can detect inappropriate content across multiple categories.

These moderation tools can be used individually or in combination to create a moderation system that fits your needs. For example, you can:

  • Use the banned words filter to automatically reject messages with obvious profanity
  • Use AI moderation to flag potentially problematic messages for review
  • Use the moderation queue to manually review flagged messages
Note: The available moderation features depend on your subscription plan. Basic moderation is available on all plans, while more advanced features are available on Pro, Premium, and Lifetime plans.

Moderation Queue

The moderation queue is a dashboard where you can review, approve, or reject messages before they appear publicly on your widget.

Accessing the Queue

To access the moderation queue:

  1. Log in to your TipPanel dashboard
  2. Click on Moderation in the side navigation
  3. Select Queue from the submenu

The moderation queue displays all messages that are pending review, with the most recent messages at the top.

Reviewing Messages

For each message in the queue, you can see:

Information Description
Message Content The text of the message
Sender Information Name and email (if provided)
Campaign The campaign the message was submitted to
Timestamp When the message was submitted
Payment Amount The amount paid (if applicable)
Flag Reason Why the message was flagged (if applicable)

To moderate a message, you have three options:

Approve

Click the Approve button to accept the message. The message will be displayed publicly on your widget.

Reject

Click the Reject button to decline the message. The message will be removed from the queue and will not be displayed.

Flag for Review

Click the Flag button to mark the message for further review later. The message will remain in the queue.

Note: If a message was submitted with a payment, rejecting the message does not refund the payment. If you need to issue a refund, you'll need to do so through your payment processor (Stripe or PayPal).

Bulk Actions

If you have many messages to moderate, you can use bulk actions to process multiple messages at once:

  1. Select the messages you want to process by checking the checkbox next to each message
  2. Click the Bulk Actions dropdown at the top of the queue
  3. Select one of the following actions:
    • Approve Selected: Approve all selected messages
    • Reject Selected: Reject all selected messages
    • Flag Selected: Flag all selected messages for review
  4. Confirm your action in the confirmation dialog
Important: Bulk actions cannot be undone, so make sure you've selected the correct messages before confirming.

Banned Words Filter

The banned words filter automatically blocks messages containing specific words or phrases that you define. This is useful for filtering out profanity, spam, or other unwanted content.

Managing Banned Words

To manage your banned words list:

  1. Log in to your TipPanel dashboard
  2. Click on Moderation in the side navigation
  3. Select Banned Words from the submenu

On the Banned Words page, you can:

Add Banned Words
  1. Enter a word or phrase in the input field
  2. Click Add to add it to your list
  3. Repeat for each word or phrase you want to ban
Remove Banned Words
  1. Find the word or phrase in your list
  2. Click the Delete button next to it
  3. Confirm the deletion in the confirmation dialog

You can also import or export your banned words list:

Import Banned Words
  1. Click the Import button
  2. Upload a text file with one word or phrase per line
  3. Click Import to add all words to your list
Export Banned Words
  1. Click the Export button
  2. A text file with your banned words list will be downloaded
  3. You can edit this file and import it later if needed
Note: TipPanel includes a default list of common profanity that you can use as a starting point. You can enable or disable this list in the filter settings.

Filter Settings

You can customize how the banned words filter works:

Setting Description Default
Filter Action What happens when a banned word is detected:
  • Block: Automatically reject the message
  • Flag: Send the message to the moderation queue
 Block
Match Type How banned words are matched:
  • Exact: Match only exact words
  • Partial: Match words within larger words
 Exact
Case Sensitivity Whether matching is case-sensitive:
  • Sensitive: "Bad" and "bad" are different
  • Insensitive: "Bad" and "bad" are the same
 Insensitive
Use Default List Whether to use TipPanel's default profanity list Enabled

To change these settings:

  1. Go to the Banned Words page
  2. Click the Settings tab
  3. Adjust the settings as needed
  4. Click Save to apply your changes

Testing the Filter

You can test your banned words filter to see if it works as expected:

  1. Go to the Banned Words page
  2. Click the Test tab
  3. Enter a test message in the input field
  4. Click Test to check if the message would be blocked

The test will show you:

  • Whether the message would be blocked or flagged
  • Which banned words were detected in the message
  • How the words were matched (exact or partial)
Tip: Use the test feature to fine-tune your banned words list and settings before applying them to your live widget.

AI Moderation

TipPanel offers AI-powered content moderation that can detect inappropriate content across multiple categories. This feature uses OpenAI's moderation API to analyze messages and identify potentially problematic content.

Setting Up AI Moderation

To set up AI moderation:

  1. Log in to your TipPanel dashboard
  2. Click on Moderation in the side navigation
  3. Select AI Moderation from the submenu
  4. Toggle the Enable AI Moderation switch to ON
  5. Enter your OpenAI API key in the input field (if not already set in your profile)
  6. Configure the moderation categories and sensitivity (see below)
  7. Click Save to apply your settings
Note: AI moderation requires an OpenAI API key, which you can get from the OpenAI website. This feature is available on Premium and Lifetime plans only.

Moderation Categories

The AI moderation system can detect content in the following categories:

Category Description Default
Hate Content that expresses, incites, or promotes hate based on race, gender, ethnicity, religion, nationality, sexual orientation, disability status, or caste Enabled
Harassment Content that expresses, incites, or promotes harassing language towards any target Enabled
Self-Harm Content that promotes, encourages, or depicts acts of self-harm, such as suicide, cutting, and eating disorders Enabled
Sexual Content meant to arouse sexual excitement, such as the description of sexual activity, or that promotes sexual services (excluding sex education and wellness) Enabled
Violence Content that promotes or glorifies violence or celebrates the suffering or humiliation of others Enabled

You can enable or disable each category based on your moderation needs. For example, if you're running a support widget for a mental health website, you might want to disable the Self-Harm category to allow users to discuss these topics.

Sensitivity Settings

For each moderation category, you can set the sensitivity level:

Low Sensitivity

Only flags content with a high probability of violating the category. May miss some borderline content but has fewer false positives.

Medium Sensitivity

Balanced approach that flags content with a moderate probability of violating the category. Default setting for most categories.

High Sensitivity

Flags content with even a low probability of violating the category. May have more false positives but catches more borderline content.

You can also set what happens when the AI detects content in a category:

Block

Automatically reject messages that violate the category

Flag

Send messages that violate the category to the moderation queue for manual review

Tip: Start with medium sensitivity and adjust based on your experience. If you're getting too many false positives, lower the sensitivity. If inappropriate content is getting through, increase the sensitivity.

Moderation Settings

You can configure global moderation settings that apply to all messages:

Setting Description Default
Pre-Moderation Whether all messages require approval before being displayed:
  • Enabled: All messages go to the moderation queue
  • Disabled: Only flagged messages go to the queue
 Disabled
Payment Bypass Whether messages with payments bypass moderation:
  • Enabled: Paid messages skip moderation
  • Disabled: All messages are moderated
 Disabled
Moderation Order The order in which moderation tools are applied:
  • Banned Words → AI: Check banned words first, then AI
  • AI → Banned Words: Check AI first, then banned words
 Banned Words → AI

To change these settings:

  1. Log in to your TipPanel dashboard
  2. Click on Moderation in the side navigation
  3. Select Settings from the submenu
  4. Adjust the settings as needed
  5. Click Save to apply your changes

Moderation Tips and Best Practices

Finding the Right Balance
  • Start with less restrictive settings and adjust as needed
  • Consider your audience and content when setting moderation rules
  • Balance between protecting your audience and allowing free expression
  • Regularly review your moderation decisions to ensure consistency
Efficient Moderation Workflow
  • Use automated tools (banned words and AI) to handle obvious cases
  • Reserve manual moderation for borderline or complex cases
  • Set aside regular time for moderation to avoid backlogs
  • Consider assigning moderation duties to team members if you have a high volume of messages
Refining Your Banned Words List
  • Start with the default list and add words specific to your context
  • Consider cultural and linguistic variations of problematic terms
  • Be aware of "leetspeak" and other obfuscation techniques (e.g., "a$$" instead of "ass")
  • Regularly update your list based on new trends or issues you observe
Optimizing AI Moderation
  • Adjust sensitivity levels based on your experience with false positives/negatives
  • Use "Flag" instead of "Block" for categories where context matters
  • Combine AI moderation with banned words for the most effective coverage
  • Monitor your OpenAI API usage to manage costs
Pro Tip: The most effective moderation strategy combines automated tools with human judgment. Use banned words and AI moderation to handle the bulk of problematic content, and reserve your time for reviewing edge cases and making nuanced decisions.
Analytics Next: Integrations