Click Here For More Support
Page cover image

Enrich Generated Responses

There are two ways to enrich your generative responses with images, buttons, quick replies, and more.

Add Enrichments Using Input Director

Add Enrichments Using Documented Data

Use one option based on the type of content you want to enrich:

Input Director

Perfect for adding enrichments to web-scraped / feed content

If you need to add training for very specific user inputs

Document

Ideal for incorporating content that is not available on your website / feed

Add Enrichments Using Input Director

To add a new enrichment for a generated response using Input Director, follow the steps below:

  1. In the Satisfi Labs dashboard, go to Studio -> NLP Manager -> Responses

  2. Click "+" in the top right of the Responses Dashboard.

  3. Create your new enrichment by adding the following fields:

Field
Details

Volume Name

Your Volume Name.

Channel Name

If the volume name is not selected first, no channels will appear.

Language

English is the only supported language at this time.

Response Name

Response Names should be all lowercase, with no spaces or special characters except for underscores. Example: enrichment_add_new_user

Content Group

The default group is General; however, you can use this field to help organize your responses in whatever works best for you.

Content Type

Select Enrichment

  1. Add the enrichments including any relevant buttons, quick replies, or rich media.

Don't add any text copy when you add enrichments using the Input Director. Our platform will use already scraped web import or feed import copy to generate responses.

  1. Select the toggle labeled Input Training and begin adding sample inputs (minimum three).

Input Training is VERY important. The quality and breadth of your inputs will determine how well training works for your enrichment.

  • We recommend having a total of 10-15 inputs per Enrichment. You must manually write at least 3 chat-like inputs before generating additional inputs.

  • Add full sentences/questions for Input Training

  1. Select the Generate Inputs button to add additional training.

  2. Delete any undesired sample inputs by selecting the toggle to the left of the input and clicking Delete Selected Inputs.

  3. Press Save New & Publish and test the new enrichment.

Add Enrichments Using Documented Data

In addition to adding a new enrichment response name, you can also add enrichments directly to documented data! To add a generative enrichment to an existing document response:

  1. While in the Satisfi Dashboard, go to Studio -> NLP Manager -> Responses

  2. Locate the document data you'd like to add an enrichment to

  3. Click the pencil icon to start editing content

  4. Add the desired enrichment using the following instructions:

Example of a Button Enrichment Added Directly in Documented Data

Below the content where you want to show the button, add an optional button label and the URL.

Example:

Label: Learn More!
URL: https://www.satisfilabs.com

If no button label accompanies a URL, your system will generate a relevant button label!

For a URL, please include "https://" before the desired URL link.

Add Inline Click Actions in Documented Data

Inline Click Actions are designed to make conversational flows cleaner and more interactive. This tool gives you the ability to embed clickable links inside messages that trigger actions and messages.

Clicking an inline link instantly sends a predefined message to the bot, just like a typed response. This is great for surfacing FAQs, creating dynamic responses, or routing within a generative response without the need for buttons.

How to Use

Inline Click Actions are powered by a special URL with three parameters:

1

Create a URL

Inline Click Actions are powered by a special URL with three parameters:

Base URL

Example: https://satis.fi/action/

This is the static endpoint used to process the action. This text will NOT change.

Name

Example: ?name=submit_ticket

A unique ID for the click action. This helps track usage or group-related actions.

Postback

Defines what the system should send when the link is clicked. There are two options:

  • Prewritten Response: Points link to a specific prewritten response record.

Example: &postback=response_name=buy_tickets

  • Free Text: Acts like input text and may trigger prewritten or generated responses.

Example: &postback=I+want+to+buy+tickets

Reply Text

Example: reply_text=Submit+Ticket!

The message shown in the user chat window when they click the link.

2

Apply the URL to Documented Data

Add the URL to your documented data within the relevant data chunk!

FAQs

If I add a button directly in documented data and I don't add a label, what does the button say?

If no button label accompanies a URL, your system will generate a relevant button label!

When should I use Input Director vs Documents when adding new enrichments?

Input Director:

  • Allows you to add buttons, quick replies, images, and gifs

  • Use Input Director to add enrichments to web-scraped / feed content

  • Training for specific user inputs

  • If the topic is very broad and lots of different generative content could be used to answer it

Document:

  • Allows you to add buttons, images, gifs

  • Ideal for incorporating content that is not available on your website / feed

  • Takes less time to add an enrichment when adding content in a documented format

Do I need to put generative enrichments in documented data next to the relevant content? Or can I add it anywhere in the response name's data?

It is important that generative enrichments always appear alongside the content where the user wants them to appear. URLs without any generative content may not be surfaced.

Last updated

Was this helpful?