The type of question or prompt used will depend on the user’s needs and the capabilities of the AI system. Effective question design is important to ensure that users get the information they need and that AI systems can provide accurate and relevant responses. In general, queries can be closed- or open-ended. Both of these types of queries have benefits and disadvntages
Closed-ended queries:
A closed-ended query is a type of question that can be answered with a simple “yes” or “no” response, or with a specific piece of information such as a number, date, or name. Closed-ended questions are typically used to elicit specific information or to confirm or deny a particular assumption. For example, in a customer service chatbot conversation, a closed-ended question might be: “Did you receive your order on time?” The answer to this question is either “yes” or “no.” Another example of a closed-ended question might be: “What is your date of birth?” In this case, the user would provide a specific date as a response.
Example: “What is the capital city of Kenya?”
Use closed-ended queries when you need a direct answer, but verify factual answers from reliable sources. Closed-ended questions are useful in conversational AI because they can help guide the conversation and provide a clear path forward. However, they can also limit the scope of the conversation and prevent users from providing more detailed or nuanced answers.
Open-ended queries:
An open-ended query is a question or statement that allows the user to provide a detailed and personalized response. These types of prompts are designed to give the user more freedom and to elicit a more creative or thoughtful response. Examples of open-ended prompts include “What do you think about X?” or “Tell me about your experience with Y.
Example: “What are three ways a teacher can use AI to support learners without replacing their own thinking?”
Use open-ended queries for brainstorming, learning, writing and exploring options. Open-ended queries allow the user to express themselves more fully, which can be especially helpful in situations where more detail is needed.
Other types of queries:
Guided queries:
A guided query is a type of search or query that is designed to help users find relevant information by providing a series of prompts, suggestions, or choices to guide the search process.
In a guided query, the system provides users with a structured and intuitive interface to help them narrow down the search criteria and find the information they are looking for. For example, a guided query for finding a hotel might ask questions such as the location, price range, amenities, and rating. The user can then select or input their preferences for each question, and the system will filter the results to show hotels that match those criteria.
Example: “Ask me one question at a time and help me create a study plan.”
By providing a structured interface and guiding users through the search process, guided queries can help reduce the cognitive load and make the search experience more efficient and effective. They are useful for complex tasks where the user needs help breaking down the process into smaller steps.
Instructional queries
These ask the AI to perform a task.
Example: “Summarise this article in five bullet points for a non-technical reader.”
Comparative queries
These ask the AI to compare options.
Example: “Compare ChatGPT, Microsoft Copilot and Google Gemini for a small business owner.”
Critical queries
These ask the AI to challenge, review or test an idea.
Example: “Review this proposal and identify weak assumptions, missing evidence and possible risks.”