Getting more out of Microsoft 365 Copilot with purposeful prompts

Editor’s note: This is the fifth video in our ongoing series that showcases our in-house experts using real-world scenarios to explore the transformative capabilities of Microsoft 365 Copilot.

Robert Gates likens Microsoft 365 Copilot to a “calculator for words.”

Gates, a partner-focused solution architect on the Global Partner Team at Microsoft, explains how to harness the full potential of Copilot through the art of purposeful prompting. He’s been helping partners across the Americas develop best practices around Copilot since its initial launch. His ongoing experience with the tool has revealed that effective use of Copilot goes far beyond simply issuing commands—it’s about crafting prompts with intention and clarity.

Press play to hear Gates share how crafting prompts based on the desired outcome can make Microsoft 365 Copilot a strategic partner.

A calculator for words

Just as a calculator requires the right inputs to produce meaningful results, Copilot needs well-constructed prompts to deliver valuable outputs.

“If the problem isn’t clearly defined, Copilot can’t generate a useful solution,” Gates says.

This is where the power of purposeful prompting comes into play.

To create effective prompts for Copilot, goals must be clearly defined. This sounds easy but can be difficult to get right. One of the key challenges Gates has observed in customer interactions is a disconnect between the data requested and the actual goal. For instance, asking Copilot to retrieve 50,000 rows of data might seem like a reasonable request, but it often misses the mark.

The real question is: What do you want to achieve with that data?

Instead of focusing on volume, Gates advises users to zero in on the outcome they want—whether it’s summarizing key insights or generating a specific report for leadership. This shift in focus can dramatically improve the quality and relevance of the responses.

“A lot of the work is defining your goal and outcome in the prompt. Sometimes you may have to work backward because Copilot doesn’t necessarily have access to the data or the resources. Focus in on the final output,” Gates says.

Crafting purposeful prompts

To get the best results from Copilot, Gates emphasizes the importance of a well-defined prompt structure. A successful prompt should include the following elements:

• Goal definition: Clearly articulate the goal of your request in your prompts. What specific output are you seeking? For example, “I need a summary of key trends from this dataset for my leadership team.”
• Context and source information: Provide relevant context that helps Copilot understand the scope and nature of the task. Mention any specific sources or data points that should be considered in generating the output.
• Expectations: Set clear expectations for the format and depth of the response. This helps Copilot tailor its output to meet your needs more precisely.
• Persona prompting: Consider nudging Copilot with a specific persona or role. For example, “As an expert analyst, summarize the key findings from this data.” This can help refine the tone and depth of the response, making it more aligned with your requirements.

By defining the outcome you want, providing context, and guiding Copilot with clear expectations, you can unlock the full potential of this powerful tool—transforming it from a basic query processor into a strategic partner in their work.

Thank you for watching Gates’s presentation! We hope it helps you get started with your Copilot training and adoption efforts at your company.