Our Response Types
Prewritten Responses
Responses that are written in advance and surfaced to users exactly as they are written.
Enriched Responses
Responses generated by the LLM model, enhanced with buttons, quick replies, and multimedia elements (images, videos), prompting users to take action.
Generated Responses
Text responses created by the LLM model using your website and documented source data.
Documented Data
Designated within the responses library with a response name starting with “doc_”. Data is pulled from uploaded documents in the dashboard to generate responses. This data can be adjusted directly within the dashboard.
Web-Scraped/URL Data
Designated within the responses library with a response name starting with “url_”. The data is directly pulled from a website to generate responses. To refresh this data, you will need to rescrape the URL.
Want to understand better how you can add responses or update your content? Explore the specific response-type pages to learn more!
Which Response Type to Use?
User Question: Can I bring a clear bag to the game?
Response Type We Recommend: Generated Response with Enrichment
Why We Recommend: Since you can scrape this information from your FAQ page and there is no need for a response that is prewritten in advanced, we recommend a personalized, generated response. We also recommend adding an enrichment, which in this case would be a button to the FAQ page so end users can be redirected for more information.
How to Identify the Source of Content Used for Generating a Response
Search for the information using keyword phrases in NLP Manager in the Dashboard
Choose the following filters:
Filter the response library to "All" and search topic keywords
Filter the response library to "Response Name" and search topic keywords
Filter the response library to "Content" and search topic keywords
Your Guide To Take Action In Case The Response is Unfavorable
Last updated