Cloudscape Design System
  • Get started
  • Foundation
  • Components
  • Patterns
  • Demos
  • Gen AI
  • GitHub 
Cloudscape Design System
  • Get started
  • Foundation
  • Components
  • Patterns
  • Demos
  • Gen AI
  • GitHub 
  • About

Gen AI

  • AI tools
    1
  • LLMs.txt files

  • Foundation
    3
  • Visual affordance
  • Colors
  • Iconography

  • Patterns
    15
  • Pattern abstraction
  • Generative AI chat
  • Ingress
  • Generative AI output label
  • Generative AI loading states
  • Progressive steps
  • Follow-up questions
  • User authorized actions
  • Variables
  • Shortcut menus
  • Response regeneration
  • Artifact previews
  • Conversational history
  • Support prompts
  • Thinking

  • Components
    6
  1. Cloudscape Design System
    • Gen AI: Patterns
    • Follow-up questions
    1. Cloudscape Design System
    2. Gen AI: Patterns
    3. Follow-up questions

    Follow-up questions

    Ways to request additional information from users to help generate more accurate and relevant responses.

    On this page

    1. Key UX concepts
    2. Criteria table
    3. Common use cases
    4. General guidelines
    5. Writing guidelines
    6. Accessibility guidelines

    About this page

    Published: July 16, 2025

    Did this page help you?

    Your feedback helps us improve our documentation.
    Provide additional feedback

    Key UX concepts

    Gather context at the right time

    Decide whether to gather information before showing output or during the interaction. For high-impact decisions or when confidence is low, collect details first. When multiple questions are needed, collecting them sequentially reduces cognitive load and keeps each decision focused. For other scenarios, start with an initial response and ask follow-up questions to refine results.

    Streamline user choices

    Provide clear options for users to proceed, skip, or exit at each step of the interaction. This ensures users feel in control of the experience and reduces friction. When structured options don’t meet the user’s needs, ensure they can always access the prompt input to provide open-ended context in their own words.

    Guide users to necessary information

    Help users access information that isn't readily available. Provide links or instructions on where to find required data when appropriate. This ensures users can provide accurate information even when it's not immediately at hand.

    Criteria table

    Information type

    Input method

    Examples

    Clarifying user intent

    Present sequential predefined choices in a dismissible dialog docked above prompt input.

    Context is needed upfront to generate an accurate response. Multiple back-and forth exchanges would otherwise be required.

    Guiding a conversation


    Present recommended inputs with support prompt group.

    The initial response applies broadly to multiple situations. A follow-up selection narrows it to the user's specific case. There are up to five valid choices.

    Common use cases

    Clarifying user intent

    When users provide ambiguous or incomplete information, or when a request requires additional information to generate an accurate response, a dismissible dialog collects structured input upfront, reducing back-and-forth exchanges in the thread.

    Guiding a conversation

    When an initial response applies broadly but a follow-up selection can tailor it to the user's specific situation, a support prompt group presents a constrained set of options after the response.

    Generative AI assistant

    JD
    I need to choose between AWS Fargate and ECS on EC2 for my containerized application.

    Both are great options for running containers. Here are the key differences:

    AWS Fargate:

    1. Serverless - no need to manage infrastructure
    2. Pay only for resources used by containers
    3. Easier to get started and maintain

    ECS on EC2:

    1. More control over underlying infrastructure
    2. Better cost optimization for consistent workloads
    3. Access to instance-level customization
    Sources
    [1] Simplify compute management with AWS Fargate 
    [2] What is Amazon EC2? 

    To provide a more tailored recommendation, what's your primary concern when choosing a container orchestration solution?

    General guidelines

    Do

    • Use structured input methods to collect follow-up information instead of open-ended text prompts.
    • To avoid overwhelming the user, limit follow-up questions to 5 or fewer per workflow.

    Don't

    • Avoid requiring users to formulate free-text responses when predefined choices or structured inputs can capture the same information more efficiently.
    • Don't ask for information already provided or data that can be retrieved automatically through APIs.

    Writing guidelines

    General writing guidelines

    • Use sentence case, but continue to capitalize proper nouns and brand names correctly in context.

    • Use end punctuation, except in headers and buttons. Don’t use exclamation points.

    • Use present-tense verbs and active voice.

    • Don't use please, thank you, ellipsis (...), ampersand (&), e.g., i.e., or etc. in writing.

    • Avoid directional language.

      • For example: use previous not above, use following not below.

    • Use device-independent language.

      • For example: use choose or select not click.

    Component-specific guidelines

    Dialog title

    • Be specific and action-oriented. Dialog titles should clearly indicate the action being performed or the purpose of the dialog, often by combining an action verb with the resource or concept involved.

      • For example: Any specific constraints I should know about?

    Placeholder text

    • Use placeholder text to provide a brief example or hint of the expected input format, not as a replacement for a label.

    • Keep placeholder text concise and descriptive enough to guide the user without cluttering the input field.

    Button text

    • If the button performs the primary action indicated in the dialog title, the verb used in the button label should match the verb in the title.

    • Follow the writing guidelines for button.

    Support prompts

    • Limit support prompt labels to less than 72 characters.

      • Follow writing guidelines for support prompt group

    Accessibility guidelines

    General accessibility guidelines

    • Follow the guidelines on alternative text and Accessible Rich Internet Applications (ARIA) regions for each component.

    • Make sure to define ARIA labels aligned with the language context of your application.

    • Don't add unnecessary markup for roles and landmarks. Follow the guidelines for each component.

    • Provide keyboard functionality to all available content in a logical and predictable order. The flow of information should make sense.

    Did this page help you?

    Your feedback helps us improve our documentation.
    Provide additional feedback
    • About
    • Connect
    • Privacy 
    • Site terms 
    • © 2026, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    Made with ❤️ at AWS