Specific and Focused Query
Definition: Queries that clearly define the desired outcome or information, avoiding vague or ambiguous language.
Ineffective Example:
Query: “Tell me about fundraising.”
Output: Provides a general, unfocused response that lacks relevance to non-profits or specific causes.
Effective Example:
Query: “Identify the key challenges faced by non-profits in fundraising for environmental causes in 2024.”
Output: Provides a detailed list of challenges specific to the non-profit sector and environmental causes.
Pro-Tip: For the latest information, try https://www.perplexity.ai/, it’s a real-time engine.
Task
Queries phrased as tasks or instructions for the LLM to complete.
- Ineffective Example:
- •Query: “Tell me how to engage donors?”
- Effective Example:
- Query: “Generate a step-by-step guide, in a table format, for creating a successful donor engagement strategy for a non-profit focused on education. “
- Output: Provides a detailed, actionable guide tailored to the specific focus area.
Context
Queries that provide relevant background information or context to help the LLM understand the broader scope.
- Ineffective Example:
- •Query: “What are sentiment analysis tools?”
- Effective Example
- Query:“Our hotel chain is looking to improve guest satisfaction and identify areas for improvement. We’re interested in using sentiment analysis to analyze online customer reviews. Can you recommend some best practices for gathering and analyzing hotel guest reviews? Additionally, are there any sentiment analysis tools specifically designed for the hospitality industry? Display the results in a table.”
- This query provides a clear context:
- Industry: Hospitality
- Goal: Improve guest satisfaction and identify areas for improvement
- Specific Task: Analyze online customer reviews using sentiment analysis
- Openness to Solutions: Asks for best practices and industry-specific tools
Open-Ended
Queries that encourage exploration and creative thinking, especially for brainstorming or generating ideas.
- Ineffective Example:
- Query:
- “What are some ideas for a customer appreciation event?
Scenario: Your company is planning its annual customer appreciation event. You want to use a large language model (LLM) to generate creative ideas for the event. However, you know that open-ended queries can be more helpful than specific ones.
Instructions:
- Consider the Audience:
- Who are you trying to engage with at this event? Existing customers? Potential new customers?
- Define Goals: What are you hoping to achieve with this event? Strengthen customer relationships? Increase brand awareness? Generate leads?
- Draft an Open-Ended Query: Considering the audience and goals, write a query that encourages the LLM to explore creative ideas for the event.
Good Query Example:
“Our company is hosting a customer appreciation event for a mix of existing and potential customers working in the tech industry. We want to create a fun and engaging experience that fosters brand loyalty and generates leads. Suggest some innovative event themes, activities, or interactive elements that would appeal to this audience and achieve our goals.”
Benefits of a Good Open-Ended Query:
- Provides context for the LLM
- Focuses on specific goals
- Encourages creative exploration
- Leads to more relevant and actionable ideas