Forum Discussion
Quick fixes to boost your Copilot responses
Did Copilot disappoint you recently?
Who has never complained about Copilot responses?
Myself included.
You know that moment when you try to get a good Copilot response, and all you get is vague one? Yep. Been there.
It’s Not You, It's not Copilot either, It's Your Prompt
People tend to treat Copilot like a search engine, but Copilot isn’t a search engine. I know it feels like it should be—type in your query, get a polished answer.
Here's the deal: Copilot (and all other LLMs) work with recognizing patterns and context based on your prompts. Copilot does not retrieve information, but generates answers. Copilot relies on the context and specificity of your prompts to generate useful responses.
That means vague, lazy prompts are setting it up to give vague responses. If you want great output, you need great input. So let’s talk about how to give Copilot the kind of prompt it can actually work with.
10 Quick Fixes to Boost Your Copilot Responses
- Write clear prompts: ❌ “Make it better” → ✅ “Summarize in 3 points”
Be specific: ❌ “Write an email” → ✅ “Draft a friendly reminder email for project deadline”
Be direct: ❌ “Could you try to…” → ✅ “Generate a bulleted list of main takeaways”
- Give context: ❌ “Draft an email” → ✅ “Draft a response to the client email /<youremail> to apologyse for the delivery delays”
- Divide to conquer: ❌ “Write a report” → ✅ “Outline the report in sections: intro, analysis, and conclusion”
Choose wisely: ❌ Copilot is not equal Copilot. → ✅ Understand the differences between M365 Copilot Chat and M365 Copilot and learn when to use what.
I wrote a post about this topic already, here you can find it: Copilot Chats vsus. M365 Copilot: What's the difference?Try rephrasing: ❌ “Try again” → ✅ “Rephrase the sentence”
Sometimes, a small word change makes a big difference.
Iterate:
Unhappy?
✅ If your plan A did not work, go for plan B.
You can start a new chat, adjust your initial prompt, and try again!
Include web content when relevant:
✅ Enrich your Copilot responses with web content.
Toggle on web content like this: Click on the 3 elipses on the top, right corner of Copilot chat in Work mode.
Give it another shot!
✅ LLMs are evolving quickly. What it did not work last month, it may work today!
Want Copilot to stop disappointing you? Prompt like you mean it.
💡 Got your own favorite Copilot trick? Share it in the comments below.
5 Replies
When I use other LLMs it helps alot to provide a System Prompt and Context Prompts to tell the LLM how you want it to respond. Is there any way to provide Copilot with similar background information?
Thank you :) I often get questions from my customers why chatGPT gives so much better answers than Copilot, even with "lazy" prompts. 😀 At the same time Copilot is supposed to use the same LLM (at least for some tasks). One thing is when using it within the productivity apps, but still when using copilot chat. How do you usually explain this?
Hi Elin,
Thanks for sharing! :) Same here. It's a common challenge. I get the same question regularly.
The first thing I usually say is: "Even though both use GPT-4o, M365 Copilot Chat is not chatGPT. "
Then I build on that, focusing on the differences and advantages of M365 Copilot:
- Seamless Integration with Microsoft 365 productivity apps
- Enterprise Data Protection
- Enhanced contextual understanding, M365 Copilot is the single point of contact for work life. It knows about your meetings, your files, your conversations, and who you are building network within your organization.
It's important here to stay focused on the purpose of the tools. chatGPT is optimized for different purposes than M365 Copilot chat. M365 Copilot chat is optimized to provide clear, compliant, and contextually relevant answers.
I prompted both tools (chatGPT pro and M365 Copilot chat) to help me brainstorm about accurate ways to answer customers the question.😍 Here my favorite answer is by M365 Copilot chat. It even gave an idea for like a comic image chatGPT vsus. Copilot by stating this:
- chatGPT may feel more “free-flowing” because it’s not always bound by the same enterprise-grade constraints.
Here are the answers for your information:
***
Answer by M365 Copilot chat [web]:Why ChatGPT Might Feel "Better" Than Copilot in Some Cases
Different Contexts, Different GoalsChatGPT is optimized for open-ended, general-purpose conversations. It’s designed to be flexible, creative, and responsive to a wide range of prompts.
Copilot, especially in Microsoft 365 apps, is optimized for productivity tasks—like summarizing emails, generating documents, or analyzing spreadsheets. It’s more context-aware but also more task-focused.
Prompt Interpretation and GuardrailsCopilot often operates with stricter safety and compliance filters, especially in enterprise environments. This can make it seem more “cautious” or “limited” in how it responds.
ChatGPT may feel more “free-flowing” because it’s not always bound by the same enterprise-grade constraints.
Context Awareness in AppsIn tools like Word, Excel, or Outlook, Copilot uses live document context to generate responses. This is powerful but also means it might not respond to general prompts the same way ChatGPT does, because it’s prioritizing in-app relevance.
User Expectations and Interface DesignChatGPT’s interface encourages exploration and creativity. Copilot’s UI is often more structured, which can lead users to expect more “buttoned-up” answers.
Users may not realize that Copilot Chat (like in Edge or Windows) is closer to ChatGPT, but still tuned for productivity and safety.
Answer by chatGPT:1. Purpose and Optimization of the Models
ChatGPT: Designed primarily for open-ended, general-purpose conversations, ChatGPT is optimized to handle a wide range of topics and provide detailed responses, even to vague prompts.
Microsoft 365 Copilot Chat: Tailored for productivity and enterprise tasks, Copilot Chat focuses on integrating with Microsoft 365 applications and providing concise, task-oriented assistance. This specialization may lead to more succinct responses in general conversational contexts.
2. Context Window and Memory
ChatGPT: Offers a larger context window, allowing it to consider more information from the conversation history, which helps in generating comprehensive responses.
Microsoft 365 Copilot Chat: Has limitations on the amount of context it can process, especially when dealing with lengthy documents or extended conversations. For instance, when summarizing or referencing content, it's recommended to keep the total to a maximum of 1.5 million words or 300 pages to ensure effective performance.
Microsoft Support3. Prompt Engineering and Default Behavior
ChatGPT: Engineered to interpret and expand upon user prompts, even when they are brief or ambiguous, aiming to provide thorough answers.
Microsoft 365 Copilot Chat: May require more precise prompts to generate detailed responses, as it is designed to operate within the structured environment of enterprise applications.
4. Training and Fine-Tuning Differences
ChatGPT: Trained on a diverse dataset encompassing a broad spectrum of topics, enabling it to handle a wide array of queries effectively.
Microsoft 365 Copilot Chat: Fine-tuned with a focus on business and professional language, which may limit its versatility in general conversational scenarios.
5. Risk and Compliance Constraints
ChatGPT: Operates with fewer restrictions, allowing for more creative and expansive outputs.
Microsoft 365 Copilot Chat: Adheres to strict data privacy and compliance standards, especially within enterprise environments, which can result in more conservative and concise responses.
Thanks frankfalvey
This is yet another great article on Copilot Amelia
