> ## Documentation Index > Fetch the complete documentation index at: https://docs.getaftercare.com/llms.txt > Use this file to discover all available pages before exploring further. # Create AI Followup Question > Generates a contextual followup question based on a survey response ## OpenAPI ````yaml POST /api/v1/followups openapi: 3.0.0 info: title: Followup Generation API version: 1.0.0 description: API for generating followup questions based on survey responses contact: name: API Support email: justin@getaftercare.com license: name: Apache 2.0 url: https://www.apache.org/licenses/LICENSE-2.0.html servers: - url: https://surveys.getaftercare.com description: Production server - url: https://staging.aftercareai.com description: Staging server security: - ApiKeyAuth: [] tags: - name: Followups description: Operations related to followup question generation - name: Data Quality description: Operations related to response data quality evaluation - name: Coding description: Operations related to coding survey responses paths: /api/v1/followups: post: tags: - Followups summary: Generate a followup question description: Generates a contextual followup question based on a survey response operationId: generateFollowup parameters: - $ref: '#/components/parameters/X-Request-Encoding' - $ref: '#/components/parameters/X-Response-Encoding' requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/FollowupGenerationRequest' example: surveyName: Customer Satisfaction surveyDescription: Tracking customer satisfaction and loyalty for the XYZ Company question: How likely are you to recommend our service? questionContext: >- I want to understand specifically why a customer is convinced to recommend our service. If they are not recommending our service, I want to know why. answer: Very likely, the customer service is excellent surveyIdentifiers: surveyIdentifier: CSAT-2024-Q1 questionIdentifier: NPS-1 guidanceBehavior: Verbose responses: '200': description: Successfully generated followup question content: application/json: schema: $ref: '#/components/schemas/FollowupGenerationResponse' example: followupQuestion: >- What specific aspects of our customer service impressed you the most? responseEvaluation: completeness: 75 relevance: 90 threadId: 7a54b8da-f65d-4946-9c9a-f202e0cc29c8 responseIdentifier: '12345' '400': description: Bad request - invalid input parameters content: application/json: schema: $ref: '#/components/schemas/Error' example: code: INVALID_REQUEST message: Invalid request parameters details: question: Question is required '403': description: Forbidden - Missing or invalid API key content: application/json: schema: $ref: '#/components/schemas/Error' example: code: INVALID_API_KEY message: Invalid or missing API key '404': description: Resource not found content: application/json: schema: $ref: '#/components/schemas/Error' example: code: NOT_FOUND message: Requested resource not found '500': description: Internal server error content: application/json: schema: $ref: '#/components/schemas/Error' example: code: INTERNAL_ERROR message: An unexpected error occurred '504': description: Gateway timeout content: application/json: schema: $ref: '#/components/schemas/Error' example: code: GATEWAY_TIMEOUT message: The request timed out security: - ApiKeyAuth: [] x-codeSamples: - lang: JavaScript source: |- const response = await api.followups.generate({ surveyName: 'Customer Satisfaction', question: 'How likely are you to recommend our service?', answer: 'Very likely, the customer service is excellent', surveyIdentifiers: { surveyIdentifier: 'CSAT-2024-Q1', questionIdentifier: 'NPS-1' } }); - lang: Python source: |- response = api.followups.generate( survey_name='Customer Satisfaction', question='How likely are you to recommend our service?', answer='Very likely, the customer service is excellent', surveyIdentifiers={ 'survey_identifier': 'CSAT-2024-Q1', 'question_identifier': 'NPS-1' } ) components: parameters: X-Request-Encoding: in: header name: X-Request-Encoding schema: type: string enum: - unicode - html default: unicode required: false description: >- Specifies the encoding format of the request data. Use `html` if you are sending data that HTML entity encodes non-ASCII characters. X-Response-Encoding: in: header name: X-Response-Encoding schema: type: string enum: - unicode - html default: unicode required: false description: >- Specifies the preferred encoding format for the response data. Use `html` if you are expecting to receive data that HTML entity encodes non-ASCII characters. schemas: FollowupGenerationRequest: type: object required: - question - answer properties: surveyName: type: string description: The name of the survey as it will appear in the Aftercare platform example: Customer Satisfaction surveyDescription: type: string description: >- A description of the survey purpose and background. It will be used to give the AI more context around specific aspects of the survey (e.g. brand/product names, movie names, etc.) example: >- Quarterly survey tracking customer satisfaction and loyalty for the XYZ Company question: type: string description: The question that was asked to the respondent example: How likely are you to recommend our service? questionContext: type: string description: >- A description of what you are trying to learn from the respondent for this question. This will be used to give the AI guidance on what followup questions to ask. example: >- I want to understand specifically why a customer is convinced to recommend our service. If they are not recommending our service, I want to know why. answer: type: string description: The answer to the question provided by the respondent example: Very likely, the customer service is excellent surveyIdentifiers: type: object description: >- Optional identifiers to tie responses to a specific survey and question. If provided, Aftercare will use these to generate analytics over your survey responses. If included, both survey and question identifiers are required. See the [Aftercare Data Model](/aftercareDataModel) to learn more. required: - surveyIdentifier - questionIdentifier properties: surveyIdentifier: type: string description: >- Optional unique identifier for the survey. Aftercare will use this to build a data model to associate questions and answers with the survey. example: CSAT-2024-Q1 questionIdentifier: type: string description: >- Optional unique identifier for the question. Aftercare will use this to build a data model to track questions across a survey example: Q1 responseLanguage: type: string description: >- Optional language to use for the followup question generation. If not provided, Aftercare will default to the language it determines best matches the response. Set response language to `match_response_language` to automatically set the language of the generated follow up question to the language of the respondent's answer. example: French threadId: type: string description: >- A unique identifier for the follow-up response thread. If provided, Aftercare will follow-up questions and answers in a thread with the original question and answer. If not provided, Aftercare will automatically generate a new thread id you can use in a subsequent follow-up. example: 7a54b8da-f65d-4946-9c9a-f202e0cc29c8 responseIdentifier: type: string description: >- Optional unique identifier for a respondent's entire response to a survey. Aftercare will use this to build a data model tie together questions and answer pairs for a single response. example: '12345' guidanceBehavior: $ref: '#/components/schemas/FollowupGenerationGuidanceBehavior' FollowupGenerationResponse: type: object required: - followupQuestion - responseEvaluation properties: followupQuestion: type: string description: The generated followup question example: >- What specific aspects of our customer service impressed you the most? responseEvaluation: $ref: '#/components/schemas/ResponseEvaluation' threadId: type: string description: The unique identifier for the follow-up thread example: 7a54b8da-f65d-4946-9c9a-f202e0cc29c8 responseIdentifier: type: string description: Optional identifier for the survey response example: '12345' Error: type: object required: - code - message properties: code: type: string description: Error code example: INVALID_REQUEST message: type: string description: Error message example: Invalid request parameters details: type: object description: Additional error details example: question: Question is required FollowupGenerationGuidanceBehavior: type: string description: >- Different ways Aftercare AI will generate follow-up questions based on incomplete or irrelevant user input. * `Succinct` - The AI will not generate a follow-up question if the response is irrelevant or complete. This occurs when the relevance score is below 25 or the completeness score is above 70. This is the default behavior. * `Redirection` - The AI will return a follow-up question that redirects the user to answering the top-level question if it is irrelevant and incomplete. A follow-up question will be generated as long as the completeness score is below 70. * `Verbose` - The AI will always return a follow-up question regardless of the relevance or completeness of the response. enum: - Succinct - Redirection - Verbose default: Succinct ResponseEvaluation: type: object required: - completeness - relevance properties: completeness: type: number description: >- A number between 0 and 100 that represents how completely the response answered the topic question. The higher the score, the more complete the response. minimum: 0 maximum: 100 relevance: type: number description: >- A number between 0 and 100 that represents the relevance of the response to the topic question. The higher the score, the more relevant the response. minimum: 0 maximum: 100 securitySchemes: ApiKeyAuth: type: apiKey in: header name: X-Aftercare-Key description: API key for authentication ````