Back to library
AI
best practice

Personalize Responses Long-Term Using Copilot Memory Instructions

Program persistent instructions into Copilot to ensure brand and style consistency.

Configure the "Copilot memory" feature to store specific instructions that the AI should always follow, eliminating the need to repeat background context in every new chat thread.

Microsoft Copilot

The Scenario

You are starting a new project and want Copilot to consistently use your specific brand voice and formatting rules without being reminded in every new chat window.

Before & after

The old way

Previously, users had to re-type context or style requirements at the start of every new interaction. This repetitive manual setup could take 5–10 minutes per session.

With AI

By setting explicit memory instructions, Copilot applies your preferences instantly to every new chat. This reduces the time spent on manual persona setting to less than 1 minute.

The Prompt

Act as my memory. Whenever I ask you to draft content, I want you to always remember: 1) Use a [TONE_PREFERENCE, e.g., professional yet accessible] tone. 2) Focus on [KEY_PROJECT_OR_TOPIC]. 3) Always provide output in [SPECIFIC_FORMAT, e.g., bulleted lists followed by a summary]. Confirm you have saved these preferences.

Leveraging Copilot's memory allows you to store specific brand guidelines, tone preferences, or project contexts that persist across chat sessions.

Source

Microsoft 365 Copilot | Microsoft Community Hub
"If you spend the time to figure out the instructions that Copilot should always remember, the quality of Copilot responses will be much better."