> For the complete documentation index, see [llms.txt](https://docs.meganova.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.meganova.ai/character-studio/chat-features/message-feedback.md).

# Message Feedback

> **Note:** Provide feedback on AI responses to help improve quality. Rate messages with thumbs up or down to contribute to character and model improvement.

## Overview

Message feedback:

* **Rate** individual responses
* **Contribute** to quality metrics
* **Help** improve characters
* **Track** response quality

## Giving Feedback

### Thumbs Up 👍

Indicates a good response:

* Stayed in character
* Engaging content
* Appropriate tone
* Met expectations

### Thumbs Down 👎

Indicates a poor response:

* Broke character
* Generic or boring
* Wrong tone
* Didn't meet expectations

## How to Rate

1. Hover over a message
2. Click 👍 or 👎
3. Rating saved automatically

## When to Rate

### Rate Positively When:

* Response shows strong personality
* Dialogue is engaging
* Information is accurate (for lore)
* Emotional response is appropriate
* Creativity is demonstrated

### Rate Negatively When:

* Character breaks role
* Response is generic/boring
* Says "As an AI..."
* Contradicts established facts
* Ignores context

## Feedback Uses

### Character Improvement

Feedback helps:

* Identify weak areas
* Track quality over time
* Guide improvements
* Measure changes

### Quality Metrics

Aggregate feedback contributes to:

* Character quality scores
* Benchmark data
* Creator insights

### Model Training

Anonymous feedback may:

* Improve AI models
* Enhance response quality
* Help MegaNova improve

## Viewing Feedback History

### Your Ratings

See what you've rated:

1. Open chat history
2. View feedback icons on messages
3. Filter by rating type

### Character Stats

Creators can see:

* Overall rating ratio
* Feedback trends
* Quality indicators

## Best Practices

### Be Consistent

Apply same standards:

* Don't rate based on mood
* Consider the criteria
* Be fair

### Rate Important Moments

Focus on:

* First impressions
* Character-defining moments
* Problem responses
* Exceptional responses

### Context Matters

Consider:

* What you asked
* Conversation flow
* Character definition
* Reasonable expectations

## Privacy

### What's Collected

* Rating (up/down)
* Message being rated
* Timestamp
* Anonymous identifier

### What's NOT Collected

* Personal information
* Full conversation context
* Identifying data

## See Also

* [Chat Sessions](https://docs.meganova.ai/character-studio/chat-features/chat-sessions.md) — Managing conversations
* [Quality Metrics](https://docs.meganova.ai/character-studio/testing-quality/quality-metrics.md) — Understanding quality
* [Benchmark Arena](https://docs.meganova.ai/character-studio/testing-quality/benchmark-arena.md) — Testing quality
