When composing your prompt, it's important to think about the specific dataset or tool the AI has been trained on. AI can only work with the data it has access to, so if you're not sure about the exact term or phrase needed to form your prompt, take a moment to research it.

For example, let's say you're working with Canvas, a learning management system, and you want to analyze rubrics and assignments. If you're not sure of the exact terminology used, your prompt might not return the results you're looking for. In Canvas, rubrics are "associated with" assignments, not "attached." By using the correct terminology, such as "associated", you can get a more accurate response from the AI.

Note: Review the relevant Canvas guide to ensure you're using the exact names of fields or data points you want to query.

AI performs best when the request is clear and measurable. Ambiguous terms like "passing grade" or "attendance" can mean different things to different people, which could lead to confusing or incomplete results. To ensure you get the data you need, try to be as quantitative as possible.

For example, instead of asking:

"List all students that are passing,"

You should specify:

"List all students who have a score of 70 or higher."

Similarly, terms like "attendance" can vary depending on the context. If your definition of attendance is based on assignment submissions rather than physical presence, make sure to define that in your prompt. For instance:

"Show me all students who have not submitted an assignment in the last 14 days."

By quantifying your request and defining terms, you'll help the AI narrow down the right data for you.

The more specific you are with your prompt, the better the AI can tailor its response. If you're looking for something very particular, make sure your prompt includes the necessary details to zero in on the exact data you're after.

For instance, if you're asking about a particular sub-account in Canvas, be sure to reference its exact name. If you're asking for data from a specific time, use the actual date ranges that are set within the Canvas system.

For example:

• Instead of asking: "All courses in week 2 module..."
• Ask: "All courses that are two weeks into the course start date..."

The latter helps the AI understand the context of your request, such as the course's start date, which allows it to return more accurate results.

If you start with a broader prompt, you can always narrow things down further within the same conversation. Just remember, specificity leads to better, more actionable insights.

AI chats are contextual, meaning that the conversation builds on previous prompts. If you start asking about one topic say, student grades and then immediately switch to a different topic, like inbox messages, the AI will try to link the two prompts. This can lead to confusion if you're jumping from one subject to another.

To avoid this, keep your conversation focused on a single topic. If you need to start a new topic, it's best to open a new chat. This will ensure that the AI responds accurately to each query based on its own context.

For example:

• If you ask about student grades in one chat, try to stay on that topic until you're done.
• If you switch to asking about assignments or student messages, consider starting a fresh conversation to avoid mixing contexts.

By using a clear, consistent flow in your prompts, you'll ensure that the AI stays on track and provides better results.