AI Event Feedback Analyzer: Typeform to Mattermost via AWS Comprehend

Unlock Real-time Event Insights with this AI-driven Workflow

This n8n workflow acts as an intelligent agent, automatically capturing feedback submitted through your Typeform event surveys. It then uses AWS Comprehend, a natural language processing (NLP) service, to perform sentiment analysis on the feedback. If negative sentiment is detected, the workflow instantly notifies your designated team channel in Mattermost, allowing for swift follow-up and issue resolution.

Key Features & Benefits

• Automated Feedback Collection: Seamlessly pulls new submissions from your Typeform forms as they happen.
• AI-Powered Sentiment Analysis: Leverages AWS Comprehend to accurately determine if feedback is positive, negative, or neutral, along with confidence scores.
• Targeted Alerts: Filters feedback and by default only notifies on negative sentiment, reducing noise and focusing attention where it's needed most.
• Real-time Mattermost Notifications: Delivers critical feedback directly to your team chat for immediate awareness and action.
• Streamlined Response Process: Enables quicker reactions to unhappy attendees, helping to manage reputation and improve future events.
• Easy Customization: Adapt the sentiment triggers, notification messages, and follow-up actions to fit your specific needs within n8n.

• An n8n instance (Cloud or self-hosted).
• Typeform account and API credentials.
• AWS account with IAM credentials configured for AWS Comprehend access (specifically, the comprehend:DetectSentiment permission).
• Mattermost instance with a Bot Access Token and the Channel ID for notifications.
• A Typeform form designed for event feedback, containing a text field for open-ended comments (e.g., named 'What did you think about the event?').

1. Download the n8n workflow JSON file.
2. Import the workflow into your n8n instance.
3. Configure the 'Typeform Trigger' node:
   • Select your Typeform API credentials.
   • Choose the specific Form Name or Form ID that collects your event feedback.
4. Configure the 'AWS Comprehend' node:
   • Select your AWS credentials.
   • Ensure the Text parameter correctly references the Typeform field containing the feedback. The template uses ={{$json["What did you think about the event?"]}}. Adjust this if your Typeform field has a different name/question.
5. Review the 'IF' node:
   • This node routes the workflow based on the detected sentiment.
   • The default condition is ={{$json["Sentiment"]}} String: Equal NEGATIVE.
   • You can modify this to act on other sentiments (e.g., POSITIVE, NEUTRAL, MIXED) or use sentiment scores for more nuanced logic.
6. Configure the 'Mattermost' node:
   • Select your Mattermost API credentials.
   • Enter the Channel ID of the Mattermost channel where notifications should be posted.
   • Customize the Message. The template uses You got new feedback with a score of {{$json["SentimentScore"]["Negative"]}}. Here is what it says:{{$node["Typeform Trigger"].json["What did you think about the event?"]}}. Adjust field references as needed, ensuring {{$json...}} paths correctly access AWS Comprehend output and {{$node["Typeform Trigger"].json...}} paths access Typeform data.
7. Test the workflow: Submit a test entry to your Typeform form with clearly negative text to verify that a notification appears in your specified Mattermost channel.
8. Activate the workflow to begin processing event feedback automatically.