{"id":13867,"date":"2025-09-29T08:20:22","date_gmt":"2025-09-29T07:20:22","guid":{"rendered":"https:\/\/nboldapp.com\/microsoft-fabric-copilot-studio-build-data-agents-over-onelake-analytics\/"},"modified":"2025-09-30T04:26:27","modified_gmt":"2025-09-30T03:26:27","slug":"microsoft-fabric-copilot-studio-build-data-agents-over-onelake-analytics","status":"publish","type":"post","link":"https:\/\/nboldapp.com\/fr\/microsoft-fabric-copilot-studio-build-data-agents-over-onelake-analytics\/","title":{"rendered":"Microsoft Fabric + Copilot Studio : Construire des agents de donn\u00e9es sur OneLake Analytics"},"content":{"rendered":"<p><strong><a href=\"https:\/\/nboldapp.com\/fr\/travailler-plus-intelligemment-et-non-plus-durement-tirer-parti-de-microsoft-copilot-pour-la-productivite\/\" style=\"display: inline;\">Microsoft Fabric<\/a>, <a href=\"https:\/\/techcommunity.microsoft.com\/t5\/microsoft-365-copilot\/bg-p\/Microsoft365CopilotBlog\" target=\"_blank\" style=\"display: inline;\">Copilot Studio<\/a>ives, et <a href=\"https:\/\/learn.microsoft.com\/en-us\/fabric\/onelake\/\" target=\"_blank\" rel=\"nofollow noopener noreferrer\" style=\"display: inline;\">OneLake<\/a> Analyse<\/strong> work together to simplify data analysis and enable conversational data interactions. Here&#8217;s how:<\/p>\n<ul>\n<li><strong>Microsoft Fabric<\/strong>: A unified platform combining data engineering, machine learning, and business intelligence. Built on <strong>OneLake<\/strong>, it centralizes organizational data for seamless access.<\/li>\n<li><strong><a href=\"https:\/\/nboldapp.com\/fr\/beginners-guide-what-is-microsoft-copilot-and-how-does-it-work\/\" style=\"display: inline;\">Copilot Studio<\/a><\/strong>: Lets you create AI-powered data agents that respond to natural language queries, simplifying complex analytics tasks.<\/li>\n<li><strong>Data Agents<\/strong>: These tools automate routine analytics, provide real-time insights, and integrate directly into apps like <a href=\"https:\/\/nboldapp.com\/fr\/microsoft-teams-for-public-organizations\/\" style=\"display: inline;\">Microsoft Teams<\/a> pour faciliter la collaboration.<\/li>\n<\/ul>\n<h3 id=\"key-features\" tabindex=\"-1\">Key Features:<\/h3>\n<ul>\n<li>Centralized data storage in <strong>OneLake<\/strong>.<\/li>\n<li>AI-driven insights via <strong><a href=\"https:\/\/nboldapp.com\/fr\/beginners-guide-using-ai-to-work-smarter-in-microsoft-365\/\" style=\"display: inline;\">Copilot Studio<\/a><\/strong>.<\/li>\n<li>Int\u00e9gration avec <a href=\"https:\/\/nboldapp.com\/fr\/outils-office-365-quels-outils-utiliser\/\" style=\"display: inline;\">Microsoft 365<\/a> apps for in-context analytics.<\/li>\n<\/ul>\n<p>To set up, you&#8217;ll need <strong>Microsoft Fabric capacity<\/strong> (F2 or higher), proper licensing, and secure authentication configurations. The result? Faster, more accessible decision-making embedded into your team&#8217;s daily workflows.<\/p>\n<h2 id=\"fabric-data-agents-copilot-studio-the-ultimate-ai-automation-duo-explained-microsoftfabric\" tabindex=\"-1\" class=\"sb h2-sbb-cls\">Fabric Data Agents + <a href=\"https:\/\/www.microsoft.com\/en-us\/microsoft-copilot\/microsoft-copilot-studio\" target=\"_blank\" rel=\"nofollow noopener noreferrer\" style=\"display: inline;\">Copilot Studio<\/a>: The Ultimate AI Automation Duo Explained! #microsoftfabric<\/h2>\n<p><img decoding=\"async\" src=\"https:\/\/assets.seobotai.com\/nboldapp.com\/68da34d1e3dd4bddfa56241c\/870edebc0cad1e03d8e32aabd0bc2ba1.jpg\" alt=\"Copilot Studio\" style=\"width:100%;\"><\/p>\n<p> <iframe class=\"sb-iframe\" src=\"https:\/\/www.youtube.com\/embed\/87xVhmuEzww\" frameborder=\"0\" loading=\"lazy\" allowfullscreen style=\"width: 100%; height: auto; aspect-ratio: 16\/9;\"><\/iframe><\/p>\n<h2 id=\"prerequisites-and-setup-requirements\" tabindex=\"-1\" class=\"sb h2-sbb-cls\">Prerequisites and Setup Requirements<\/h2>\n<p>Before diving into building data agents that connect Microsoft Fabric with Copilot Studio, make sure your organization meets the necessary technical and licensing requirements.<\/p>\n<h3 id=\"technical-requirements\" tabindex=\"-1\">Exigences techniques<\/h3>\n<p>To use Microsoft Fabric with Copilot Studio, your organization needs a Fabric capacity of <strong>F2 or higher<\/strong>. For reference, an <strong>F64 SKU<\/strong> costs approximately $8,409.60 per month on a pay-as-you-go plan, or around $5,002.67 per month with a one-year reservation. Smaller Fabric SKUs require every user to have a <a href=\"https:\/\/nboldapp.com\/fr\/power-bi-with-microsoft-teams-how-to-manage-reports-effectively\/\" style=\"display: inline;\">Power BI<\/a> Pro, Premium Per User, or trial license. However, with F64+ SKUs, dedicated <a href=\"https:\/\/nboldapp.com\/fr\/power-bi-with-microsoft-teams-how-to-manage-reports-effectively\/\" style=\"display: inline;\">Power BI<\/a> capabilities are included, eliminating the need for separate <a href=\"https:\/\/nboldapp.com\/fr\/power-bi-with-microsoft-teams-how-to-manage-reports-effectively\/\" style=\"display: inline;\">Power BI<\/a> licenses.<\/p>\n<p>If you want to enable Copilot functionality within Fabric, you\u2019ll need an <strong>F64 or higher SKU<\/strong>. This is essential for accessing the full conversational AI capabilities, which can involve a notable investment in higher capacity tiers.<\/p>\n<p>For Copilot Studio licensing, there are three options:<\/p>\n<ul>\n<li><strong>Pay-as-you-go<\/strong>: $0.01 per Copilot Credit with an active <a href=\"https:\/\/azure.microsoft.com\/en-us\" target=\"_blank\" rel=\"nofollow noopener noreferrer\" style=\"display: inline;\">Azure<\/a> subscription.<\/li>\n<li><strong>Copilot Credit packs<\/strong>: $200 per tenant per month for 25,000 credits.<\/li>\n<li><strong><a href=\"https:\/\/www.microsoft.com\/en-us\/microsoft-365\" target=\"_blank\" rel=\"nofollow noopener noreferrer\" style=\"display: inline;\">Microsoft 365<\/a> Copilot licenses<\/strong>: $30 per user per month.<\/li>\n<\/ul>\n<p>Once capacity and licensing are sorted, you\u2019ll need to configure authentication and tenant settings to establish secure connections.<\/p>\n<h3 id=\"authentication-and-permissions\" tabindex=\"-1\">Authentication and Permissions<\/h3>\n<p>After setting up capacity and licensing, your next step is enabling the right tenant configurations. Specifically:<\/p>\n<ul>\n<li><strong>Fabric data agent tenant setting<\/strong>: This must be activated to allow the creation and deployment of data agents in your Fabric environment.<\/li>\n<li><strong>Copilot tenant switch<\/strong>: This setting is essential for enabling Copilot functionality.<\/li>\n<\/ul>\n<p>In the <strong><a href=\"https:\/\/www.microsoft.com\/en-us\/power-platform\/products\/power-bi\" target=\"_blank\" rel=\"nofollow noopener noreferrer\" style=\"display: inline;\">Power BI<\/a> admin portal<\/strong>, enable the <strong>standalone Copilot experience<\/strong> by navigating to Tenant settings &gt; Copilot &gt; Standalone Copilot experience. This allows users to interact with Copilot independently of specific workspaces or reports.<\/p>\n<p>If your organization operates across multiple regions, you\u2019ll need to enable <strong>Cross-geo AI processing and storage<\/strong>. This ensures AI processing can occur across different data centers while adhering to data residency requirements.<\/p>\n<p>For Power BI semantic models, it\u2019s crucial to activate the <strong>XMLA endpoints<\/strong> tenant switch. This allows external applications and services to access and query your Power BI datasets, which is essential for enabling data agents to function.<\/p>\n<p>When it comes to user-level permissions in Copilot Studio, users must have one of the following:<\/p>\n<ul>\n<li>A dedicated <a href=\"https:\/\/nboldapp.com\/fr\/guide-de-lutilisateur-avance-fonctionnalites-avancees-de-microsoft-365-copilot\/\" style=\"display: inline;\">Copilot Studio user license<\/a> (requires a prepaid Copilot Credit pack subscription),<\/li>\n<li>The &quot;Copilot Studio authors&quot; role in the Power Platform admin center, or<\/li>\n<li>A <a href=\"https:\/\/nboldapp.com\/fr\/microsoft-365-copilot-vs-traditional-ai-assistants-what-makes-copilot-unique\/\" style=\"display: inline;\">Licence Microsoft 365 Copilot<\/a>.<\/li>\n<\/ul>\n<h3 id=\"initial-setup-steps\" tabindex=\"-1\">\u00c9tapes de la configuration initiale<\/h3>\n<p>To get started, you\u2019ll need to configure at least one data source. This could be a <strong>warehouse<\/strong>, <strong>lakehouse<\/strong>, <strong><a href=\"https:\/\/nboldapp.com\/fr\/customizing-microsoft-copilot-for-your-business-needs-enterprise-guide\/\" style=\"display: inline;\">Power BI semantic model<\/a><\/strong>, or <strong>KQL database<\/strong>. Without a data source containing actual data, your agents won\u2019t have anything meaningful to analyze or query.<\/p>\n<p>Make sure <strong>Fabric workspaces<\/strong> are set up with the correct user permissions. Workspace members must have roles that allow them to create, modify, and publish data agents within their assigned workspaces.<\/p>\n<p>For many organizations, starting with a pilot project using the <strong>pay-as-you-go Copilot Studio model<\/strong> is a practical way to evaluate its potential. If your organization already has <a href=\"https:\/\/nboldapp.com\/fr\/what-is-microsoft-copilot-and-how-does-it-work\/\" style=\"display: inline;\">Licences Microsoft 365 Copilot<\/a>, integration could be more cost-effective since these licenses include many of the capabilities needed for extending Copilot Studio with custom agents.<\/p>\n<h2 id=\"step-by-step-guide-building-and-connecting-data-agents\" tabindex=\"-1\" class=\"sb h2-sbb-cls\">Step-by-Step Guide: Building and Connecting Data Agents<\/h2>\n<p>With your environment set up, it&#8217;s time to create data agents that can transform how your team interacts with data. This involves three main stages: setting up a solid data foundation, building the agent itself, and integrating it into Copilot Studio for organization-wide access. Start by creating a Lakehouse, move on to developing your agent, and then connect it to Copilot Studio.<\/p>\n<h3 id=\"creating-a-lakehouse-in-microsoft-fabric\" tabindex=\"-1\">Creating a Lakehouse in <a href=\"https:\/\/www.microsoft.com\/en-us\/microsoft-fabric\" target=\"_blank\" rel=\"nofollow noopener noreferrer\" style=\"display: inline;\">Microsoft Fabric<\/a><\/h3>\n<p><img decoding=\"async\" src=\"https:\/\/assets.seobotai.com\/nboldapp.com\/68da34d1e3dd4bddfa56241c\/bdee3ec15375544518d2bf827ce52a65.jpg\" alt=\"Microsoft Fabric\" style=\"width:100%;\"><\/p>\n<p>A Lakehouse serves as the backbone of your data operations, offering scalable file storage and the metastore functionality you need for modern analytics workflows. Built on the <strong><a href=\"https:\/\/delta.io\/\" target=\"_blank\" rel=\"nofollow noopener noreferrer\" style=\"display: inline;\">Delta Lake<\/a> format<\/strong>, it ensures your data stays consistent and accessible across various tools.<\/p>\n<p>To get started, head to <a href=\"https:\/\/app.fabric.microsoft.com\/home?experience=fabric\" target=\"_blank\" rel=\"nofollow noopener noreferrer\" style=\"display: inline;\">Microsoft Fabric<\/a>, sign in, and navigate to <strong>Espaces de travail<\/strong>.<\/p>\n<p>You can either select an existing Fabric-enabled workspace or create a new one. Make sure the workspace has a licensing mode that supports Fabric capacity (like Trial, Premium, or Fabric), as this is crucial for setting up the data infrastructure.<\/p>\n<p>In your workspace, click <strong>New item<\/strong> (or choose <strong>Cr\u00e9er<\/strong> from the left menu) and find <strong>Lakehouse<\/strong> under the &quot;Store data&quot; or &quot;Data Engineering&quot; section. Once selected, name your Lakehouse descriptively. For instance, if it&#8217;s for sales data, a name like &quot;SalesAnalyticsLakehouse&quot; makes its purpose clear and aligns with your team&#8217;s workflows.<\/p>\n<p>Don&#8217;t forget to assign sensitivity labels where needed. With your Lakehouse ready, you\u2019re set to build a data agent that can tap into this resource.<\/p>\n<h3 id=\"building-and-publishing-a-fabric-data-agent\" tabindex=\"-1\">Building and Publishing a Fabric Data Agent<\/h3>\n<p>Cliquez sur <strong>+ New Item<\/strong> and search for &quot;Fabric data agent&quot; in the preview options. Once located, create it and give it a purpose-driven name, such as &quot;Sales Insights Agent&quot; or &quot;Customer Analytics Assistant.&quot;<\/p>\n<p>The OneLake catalog will display your available data sources. These agents support up to five sources from approved options. Add your data sources by selecting them and clicking <strong>Ajouter<\/strong>. Include your newly created Lakehouse along with any existing resources like warehouses, Power BI semantic models, or KQL databases.<\/p>\n<p>The agent uses Azure OpenAI Assistant APIs to turn user requests into precise, read-only queries, ensuring your data remains secure from unintended changes.<\/p>\n<p>In the <strong>Explorer<\/strong> pane, select the tables the AI should access. Use descriptive table names like &quot;SalesData&quot; or &quot;CustomerMetrics&quot; to avoid confusion. Generic names like &quot;Table1&quot; should be avoided for clarity.<\/p>\n<p>In the <strong>Data agent instructions<\/strong> pane, you can guide the AI&#8217;s behavior. You have up to 15,000 characters to provide instructions, such as which data sources to prioritize. For example, you might direct financial queries to Power BI models while sending sales-related questions to your Lakehouse.<\/p>\n<p>Adding example queries helps improve accuracy. Use the &quot;Example queries&quot; feature to provide sample question-query pairs. These examples should align with your data schema and use valid syntax to help the AI understand expected patterns.<\/p>\n<p>Test the agent with a mix of simple and complex queries, such as &quot;How many customers do we have?&quot; or &quot;What\u2019s the month-over-month growth rate for the top three product categories?&quot; Once satisfied with its performance, click <strong>Publish<\/strong>. Add a detailed description of the agent\u2019s capabilities to help team members understand its purpose.<\/p>\n<p>After publishing, you\u2019ll have two versions: a draft for ongoing updates and a published version for team access. This allows you to refine the agent without disrupting active users.<\/p>\n<h3 id=\"connecting-the-data-agent-to-copilot-studio\" tabindex=\"-1\">Connecting the Data Agent to Copilot Studio<\/h3>\n<p>With your agent published, the next step is to integrate it into Copilot Studio for a seamless analytics experience. This integration allows your team to access insights directly through <a href=\"https:\/\/nboldapp.com\/fr\/les-astuces-de-productivite-de-lai-permettent-de-gagner-du-temps-avec-microsoft-teams-et-copilot\/\" style=\"display: inline;\">Microsoft Teams<\/a>, <a href=\"https:\/\/nboldapp.com\/fr\/sharepoint-with-microsoft-teams\/\" style=\"display: inline;\">SharePoint<\/a>, or other <a href=\"https:\/\/nboldapp.com\/fr\/10-ways-to-boost-productivity-with-ai-in-microsoft-365\/\" style=\"display: inline;\">Applications Microsoft 365<\/a>.<\/p>\n<p>Your Fabric Data Agent will automatically appear as a plugin in Copilot Studio after publishing. Authentication is handled via your existing <a href=\"https:\/\/www.microsoft.com\/en-us\/security\/business\/identity-access\/microsoft-entra-id\" target=\"_blank\" rel=\"nofollow noopener noreferrer\" style=\"display: inline;\">Microsoft Entra ID<\/a> setup, ensuring data access aligns with user permissions and workspace security.<\/p>\n<p>In Copilot Studio, you can either create a new custom copilot or edit an existing one to include your Fabric Data Agent. The agent will be listed among available plugins, where you can configure its role in conversational workflows and business logic.<\/p>\n<p>Set up conversation starters like &quot;Ask about quarterly sales performance&quot; to guide users on how to interact with the agent. Additionally, configure responses to include relevant context, such as timeframes or data source details. Test the integration by having team members use the agent during their regular Microsoft 365 tasks.<\/p>\n<p>Monitor its performance, paying close attention to how it handles follow-up questions and clarifications. This ensures it operates smoothly and delivers meaningful insights.<\/p>\n<h6 id=\"sbb-itb-8be0fd2\" tabindex=\"-1\">sbb-itb-8be0fd2<\/h6>\n<h2 id=\"use-cases-workflow-automation-and-productivity-in-microsoft-teams\" tabindex=\"-1\" class=\"sb h2-sbb-cls\">Use Cases: Workflow Automation and Productivity in <a href=\"https:\/\/www.microsoft.com\/en-us\/microsoft-teams\/group-chat-software\" target=\"_blank\" rel=\"nofollow noopener noreferrer\" style=\"display: inline;\">Microsoft Teams<\/a><\/h2>\n<p><img decoding=\"async\" src=\"https:\/\/assets.seobotai.com\/nboldapp.com\/68da34d1e3dd4bddfa56241c\/33ad1d9a30b6c2d9e293fb0623cc4b55.jpg\" alt=\"Microsoft Teams\" style=\"width:100%;\"><\/p>\n<p>Data agents bring analytics directly into daily conversations and workflows. Instead of juggling multiple apps or waiting for reports, team members can access insights instantly within <a href=\"https:\/\/nboldapp.com\/fr\/microsoft-copilot-and-teams-revolutionizing-workplace-collaboration\/\" style=\"display: inline;\">Microsoft Teams<\/a> channels and chats. This integration allows for smarter decision-making and smoother processes across the organization. Below are examples of how data agents simplify analytics and empower teams such as sales and finance within the Teams environment.<\/p>\n<h3 id=\"automating-sales-analytics\" tabindex=\"-1\">Automating Sales Analytics<\/h3>\n<p>Sales teams often spend too much time compiling reports and tracking performance metrics. Data agents cut down on this manual work by providing instant access to key metrics, customer insights, and revenue data directly within Teams conversations.<\/p>\n<p>For instance, if a sales manager asks, &quot;What&#8217;s our conversion rate this quarter?&quot; in a Teams chat, the agent can immediately pull real-time figures, highlight trends, and provide context using OneLake analytics.<\/p>\n<p>Data agents also make pipeline updates more efficient during daily standups. If someone asks, &quot;Which deals are at risk of slipping this month?&quot;, the agent delivers detailed insights, including deal values, expected close dates, and recommendations based on historical data. Additionally, customer relationship details &#8211; like interaction histories and purchase trends &#8211; are integrated into conversations, helping sales reps personalize their approach and uncover upselling opportunities effortlessly.<\/p>\n<h3 id=\"real-time-team-decision-making\" tabindex=\"-1\">Real-Time Team Decision-Making<\/h3>\n<p>Project and finance teams often need quick access to data for on-the-spot decisions. Data agents make this possible by delivering analytics in real time, right within the flow of a conversation, eliminating the need for technical expertise or workflow interruptions.<\/p>\n<p>During budget planning, finance teams can query spending data, forecasts, and variance reports directly in Teams. This on-demand access reduces the need for separate review meetings and speeds up decision-making. Similarly, marketing teams can check campaign performance by simply asking, &quot;How are our social media campaigns performing this week?&quot; and instantly receive engagement metrics, conversion rates, and budget usage stats.<\/p>\n<p>Operations teams also benefit by monitoring key performance indicators during shift changes or crisis situations. They can get instant updates on production, quality, or supply chain metrics, along with actionable recommendations, ensuring that critical decisions are based on up-to-date information.<\/p>\n<h3 id=\"nbold-integration-for-better-collaboration\" tabindex=\"-1\"><a href=\"https:\/\/nboldapp.com\/fr\/\" style=\"display: inline;\">nBold<\/a> Integration for Better Collaboration<\/h3>\n<p><img decoding=\"async\" src=\"https:\/\/assets.seobotai.com\/nboldapp.com\/68da34d1e3dd4bddfa56241c\/a5639dff5705fe6c3c351df8eb2ac080.jpg\" alt=\"nBold\" style=\"width:100%;\"><\/p>\n<p><a href=\"https:\/\/nboldapp.com\/fr\/\" style=\"display: inline;\">nBold<\/a> adds another layer of efficiency by offering template-based governance for structured and compliant collaboration spaces. By automating team creation and standardizing channel structures, <a href=\"https:\/\/nboldapp.com\/fr\/\" style=\"display: inline;\">nBold<\/a> ensures that the analytics capabilities of data agents are consistently available across all workspaces.<\/p>\n<p>With nBold&#8217;s customizable templates, organizations can design workspaces tailored to specific business functions &#8211; whether for sales, marketing, or operations &#8211; so teams can quickly access the most relevant data insights for their roles. Additionally, nBold&#8217;s governance policies help maintain data security and compliance by restricting access to sensitive financial information while allowing broader sharing of general performance metrics.<\/p>\n<h2 id=\"troubleshooting-and-best-practices\" tabindex=\"-1\" class=\"sb h2-sbb-cls\">Troubleshooting and Best Practices<\/h2>\n<p>Using data agents in OneLake Analytics can sometimes present challenges related to authentication, permissions, and integration. By identifying common issues and following practical strategies, you can ensure a smoother setup and more reliable performance.<\/p>\n<h3 id=\"common-setup-issues-and-solutions\" tabindex=\"-1\">Common Setup Issues and Solutions<\/h3>\n<p>Even when you follow the setup steps, a few common problems might arise that could disrupt the integration process:<\/p>\n<ul>\n<li> <strong>Visibility Issues<\/strong>: Sometimes, data agents may not appear in Copilot Studio, even though they\u2019ve been created in Microsoft Fabric. This typically happens if the Fabric data agent hasn\u2019t been published or is still in draft mode. Make sure the agent is published and set to a &quot;Running&quot; state before attempting to connect it to Copilot Studio. <\/li>\n<li> <strong>Authentication Mismatches<\/strong>: If the accounts used in Microsoft Fabric and Copilot Studio don\u2019t match, connection failures can occur. To avoid this, ensure you\u2019re signed in with the same account across both platforms, and that both services are operating under the same Microsoft 365 tenant. <\/li>\n<li> <strong>Permission Errors<\/strong>: These arise when users lack the necessary access rights to the underlying data sources. For example:\n<ul>\n<li>Power BI semantic models require &quot;Build&quot; permission (along with basic Read access) since data agents generate queries that need elevated permissions.<\/li>\n<li>Lakehouse and warehouse connections both require &quot;Read&quot; (SELECT) permissions.<\/li>\n<li>For KQL databases, users must have the &quot;Reader&quot; role.<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n<h3 id=\"best-practices-for-data-agents\" tabindex=\"-1\">Best Practices for Data Agents<\/h3>\n<p>To avoid these common pitfalls, follow these best practices:<\/p>\n<ul>\n<li> <strong>Grant Minimal Permissions<\/strong>: Assign only the permissions necessary for data querying to reduce security risks. <\/li>\n<li> <strong>Set Sharing Permissions Based on Roles<\/strong>:\n<ul>\n<li><em>No permission<\/em>: For users who simply need query access.<\/li>\n<li><em>View details<\/em>: For stakeholders who need visibility into the agent setup but shouldn\u2019t make changes.<\/li>\n<li><em>Edit and view details<\/em>: For team members responsible for creating and maintaining the agents.<\/li>\n<\/ul>\n<\/li>\n<li> <strong>Leverage <a href=\"https:\/\/nboldapp.com\/fr\/a-deep-dive-into-microsoft-365-copilot-wave-2-spring-release\/\" style=\"display: inline;\">Microsoft Entra<\/a> ID Integration<\/strong>: Use conditional access policies to enhance security. These policies can enforce multifactor authentication, restrict access to devices enrolled in Intune, and limit access based on user locations or specific IP ranges. <\/li>\n<li> <strong>Use Workspace Security<\/strong>: Treat workspace security as the main boundary for data stored in OneLake. Permissions should be managed through workspace role assignments to maintain control over access. <\/li>\n<\/ul>\n<h2 id=\"key-takeaways\" tabindex=\"-1\" class=\"sb h2-sbb-cls\">Principaux enseignements<\/h2>\n<p>Using Microsoft Fabric and Copilot Studio to build data agents is changing how teams interact with data in OneLake. This integration simplifies complex data insights, embedding them directly into everyday workflows. The result? Faster, more accessible decision-making driven by data across the organization.<\/p>\n<p>To ensure smooth deployment of data agents, it&#8217;s essential to establish a solid technical foundation. This includes setting up proper permissions, authentication, and secure workspace configurations. Balancing strong workspace security with role-based permissions ensures both protection and operational flexibility.<\/p>\n<p>The true value shines when data agents are seamlessly integrated into Microsoft Teams workflows. Imagine sales teams automating analytics reports, leadership accessing real-time insights for decisions, and collaborative efforts becoming more informed &#8211; all without requiring deep technical know-how. This underscores the importance of a secure, well-thought-out technical setup.<\/p>\n<p>For organizations using nBold&#8217;s collaboration templates, data agents can be embedded into standardized team templates. This means every new project team created from a template automatically gains access to relevant data insights, promoting consistent, data-driven practices across all collaboration spaces.<\/p>\n<p>Effective troubleshooting hinges on adhering to best practices, especially when it comes to permissions and maintaining account consistency.<\/p>\n<p>Looking back on the process of building and integrating data agents, the strategic edge lies in enabling broad, secure access to data while maintaining strict security. Business users benefit from self-service analytics powered by natural language interactions, while IT teams retain control through strong permission structures and governance policies.<\/p>\n<h2 id=\"faqs\" tabindex=\"-1\" class=\"sb h2-sbb-cls\">FAQ<\/h2>\n<h3 id=\"how-do-microsoft-fabric-and-copilot-studio-work-together-to-improve-data-analytics\" tabindex=\"-1\" data-faq-q>How do Microsoft Fabric and Copilot Studio work together to improve data analytics?<\/h3>\n<p>Microsoft Fabric and Copilot Studio work together to make data analytics more efficient and user-friendly by combining <strong>OneLake<\/strong>, a unified data lake, with AI-driven tools. This integration helps businesses centralize their data, automate repetitive tasks, and turn raw information into meaningful insights with ease.<\/p>\n<p>Utilisation <strong>Copilot Studio<\/strong>, teams can take advantage of natural language queries, get smart suggestions, and simplify workflows. By cutting down on manual work, these tools not only improve collaboration but also speed up decision-making and open the door to new ideas, making analytics easier to use across the board.<\/p>\n<h3 id=\"how-do-data-agents-in-microsoft-teams-help-improve-team-workflows\" tabindex=\"-1\" data-faq-q>How do data agents in Microsoft Teams help improve team workflows?<\/h3>\n<p>Data agents in Microsoft Teams are game-changers for team workflows. They take care of <strong>repetitive tasks<\/strong>, provide instant access to shared knowledge, and support real-time data analysis. This means less manual work and more time saved for what truly matters.<\/p>\n<p>These agents eliminate bottlenecks by breaking down information silos, giving teams immediate access to crucial data. This not only boosts collaboration but also empowers teams to make quicker, data-driven decisions. The outcome? A smoother, more productive, and efficient workspace.<\/p>\n<h3 id=\"what-are-the-costs-and-licensing-requirements-for-using-microsoft-fabric-and-copilot-studio-together\" tabindex=\"-1\" data-faq-q>What are the costs and licensing requirements for using Microsoft Fabric and Copilot Studio together?<\/h3>\n<p>The cost of using <strong>Microsoft Fabric<\/strong> et <strong>Copilot Studio<\/strong> varies depending on the capacity and usage you choose. Microsoft Fabric pricing starts at <strong>$262.80 per month<\/strong> for F2 capacity and can go up to <strong>$2,102.40 per month<\/strong> for F16 capacity.<\/p>\n<p>Pour <strong>Copilot Studio<\/strong>, pricing is based on a prepaid credit system. It begins at <strong>$200 per tenant per month<\/strong>, which includes 25,000 credits. If your usage exceeds this, you can purchase additional capacity packs to meet your needs.<\/p>\n<p>En plus de cela, <strong>Microsoft 365 Copilot<\/strong> costs <strong>$30 par utilisateur et par mois<\/strong>. Copilot Studio may also require a separate licensing fee, estimated at around <strong>$36,000<\/strong>, along with potential charges based on message volume. It&#8217;s essential to assess your business requirements carefully to configure these tools in the most cost-effective way.<\/p>\n<h2>Articles de blog connexes<\/h2>\n<ul>\n<li><a href=\"\/fr\/microsoft-365-copilot-vs-traditional-ai-assistants-what-makes-copilot-unique\/\" style=\"display: inline;\">Microsoft 365 Copilot vs. Traditional AI Assistants: What Makes Copilot Unique?<\/a><\/li>\n<li><a href=\"\/fr\/customizing-microsoft-copilot-for-your-business-needs-enterprise-guide\/\" style=\"display: inline;\">Customizing Microsoft Copilot for Your Business Needs (Enterprise Guide)<\/a><\/li>\n<li><a href=\"\/fr\/10-ways-to-boost-productivity-with-ai-in-microsoft-365\/\" style=\"display: inline;\">10 Ways to Boost Productivity with AI in Microsoft 365<\/a><\/li>\n<li><a href=\"\/fr\/a-deep-dive-into-microsoft-365-copilot-wave-2-spring-release\/\" style=\"display: inline;\">Une plong\u00e9e en profondeur dans Microsoft 365 Copilot Wave 2 Spring Release<\/a><\/li>\n<\/ul>\n<p><script async type=\"text\/javascript\" src=\"https:\/\/app.seobotai.com\/banner\/banner.js?id=68da34d1e3dd4bddfa56241c\"><\/script><\/p>","protected":false},"excerpt":{"rendered":"<p>Explore how Microsoft Fabric and Copilot Studio streamline data analysis, enabling teams to leverage AI-driven insights within their workflows.<\/p>","protected":false},"author":11,"featured_media":13866,"comment_status":"closed","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"content-type":"","inline_featured_image":false,"footnotes":""},"categories":[745],"tags":[],"class_list":["post-13867","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-teams"],"featured_image_src":"https:\/\/nboldapp.com\/wp-content\/uploads\/2025\/09\/image_fd8c8d3c2de46016c6ef2f849a0f73af.jpeg","author_info":{"display_name":"Alexandre Cipriani","author_link":"https:\/\/nboldapp.com\/fr\/author\/alexandre-cipriani\/"},"yoast_head":"<!-- This site is optimized with the Yoast SEO Premium plugin v27.2 (Yoast SEO v27.2) - https:\/\/yoast.com\/product\/yoast-seo-premium-wordpress\/ -->\n<title>Microsoft Fabric + Copilot Studio: Build Data Agents over OneLake Analytics - nBold<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/nboldapp.com\/fr\/microsoft-fabric-copilot-studio-build-data-agents-over-onelake-analytics\/\" \/>\n<meta property=\"og:locale\" content=\"fr_FR\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Microsoft Fabric + Copilot Studio: Build Data Agents over OneLake Analytics\" \/>\n<meta property=\"og:description\" content=\"Explore how Microsoft Fabric and Copilot Studio streamline data analysis, enabling teams to leverage AI-driven insights within their workflows.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/nboldapp.com\/fr\/microsoft-fabric-copilot-studio-build-data-agents-over-onelake-analytics\/\" \/>\n<meta property=\"og:site_name\" content=\"nBold\" \/>\n<meta property=\"article:published_time\" content=\"2025-09-29T07:20:22+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2025-09-30T03:26:27+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/nboldapp.com\/wp-content\/uploads\/2025\/09\/image_fd8c8d3c2de46016c6ef2f849a0f73af.jpeg\" \/>\n\t<meta property=\"og:image:width\" content=\"1536\" \/>\n\t<meta property=\"og:image:height\" content=\"1024\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\n<meta name=\"author\" content=\"Alexandre Cipriani\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@nboldhq\" \/>\n<meta name=\"twitter:site\" content=\"@nboldhq\" \/>\n<meta name=\"twitter:label1\" content=\"\u00c9crit par\" \/>\n\t<meta name=\"twitter:data1\" content=\"Alexandre Cipriani\" \/>\n\t<meta name=\"twitter:label2\" content=\"Dur\u00e9e de lecture estim\u00e9e\" \/>\n\t<meta name=\"twitter:data2\" content=\"14 minutes\" \/>\n<!-- \/ Yoast SEO Premium plugin. -->","yoast_head_json":{"title":"Microsoft Fabric + Copilot Studio: Build Data Agents over OneLake Analytics - nBold","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/nboldapp.com\/fr\/microsoft-fabric-copilot-studio-build-data-agents-over-onelake-analytics\/","og_locale":"fr_FR","og_type":"article","og_title":"Microsoft Fabric + Copilot Studio: Build Data Agents over OneLake Analytics","og_description":"Explore how Microsoft Fabric and Copilot Studio streamline data analysis, enabling teams to leverage AI-driven insights within their workflows.","og_url":"https:\/\/nboldapp.com\/fr\/microsoft-fabric-copilot-studio-build-data-agents-over-onelake-analytics\/","og_site_name":"nBold","article_published_time":"2025-09-29T07:20:22+00:00","article_modified_time":"2025-09-30T03:26:27+00:00","og_image":[{"width":1536,"height":1024,"url":"https:\/\/nboldapp.com\/wp-content\/uploads\/2025\/09\/image_fd8c8d3c2de46016c6ef2f849a0f73af.jpeg","type":"image\/jpeg"}],"author":"Alexandre Cipriani","twitter_card":"summary_large_image","twitter_creator":"@nboldhq","twitter_site":"@nboldhq","twitter_misc":{"\u00c9crit par":"Alexandre Cipriani","Dur\u00e9e de lecture estim\u00e9e":"14 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/nboldapp.com\/microsoft-fabric-copilot-studio-build-data-agents-over-onelake-analytics\/#article","isPartOf":{"@id":"https:\/\/nboldapp.com\/microsoft-fabric-copilot-studio-build-data-agents-over-onelake-analytics\/"},"author":{"name":"Alexandre Cipriani","@id":"https:\/\/nboldapp.com\/#\/schema\/person\/ffd46719c510e6ee95af907cd2c2f985"},"headline":"Microsoft Fabric + Copilot Studio: Build Data Agents over OneLake Analytics","datePublished":"2025-09-29T07:20:22+00:00","dateModified":"2025-09-30T03:26:27+00:00","mainEntityOfPage":{"@id":"https:\/\/nboldapp.com\/microsoft-fabric-copilot-studio-build-data-agents-over-onelake-analytics\/"},"wordCount":2911,"publisher":{"@id":"https:\/\/nboldapp.com\/#organization"},"image":{"@id":"https:\/\/nboldapp.com\/microsoft-fabric-copilot-studio-build-data-agents-over-onelake-analytics\/#primaryimage"},"thumbnailUrl":"https:\/\/nboldapp.com\/wp-content\/uploads\/2025\/09\/image_fd8c8d3c2de46016c6ef2f849a0f73af.jpeg","articleSection":["Teams"],"inLanguage":"fr-FR"},{"@type":"WebPage","@id":"https:\/\/nboldapp.com\/microsoft-fabric-copilot-studio-build-data-agents-over-onelake-analytics\/","url":"https:\/\/nboldapp.com\/microsoft-fabric-copilot-studio-build-data-agents-over-onelake-analytics\/","name":"Microsoft Fabric + Copilot Studio: Build Data Agents over OneLake Analytics - nBold","isPartOf":{"@id":"https:\/\/nboldapp.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/nboldapp.com\/microsoft-fabric-copilot-studio-build-data-agents-over-onelake-analytics\/#primaryimage"},"image":{"@id":"https:\/\/nboldapp.com\/microsoft-fabric-copilot-studio-build-data-agents-over-onelake-analytics\/#primaryimage"},"thumbnailUrl":"https:\/\/nboldapp.com\/wp-content\/uploads\/2025\/09\/image_fd8c8d3c2de46016c6ef2f849a0f73af.jpeg","datePublished":"2025-09-29T07:20:22+00:00","dateModified":"2025-09-30T03:26:27+00:00","breadcrumb":{"@id":"https:\/\/nboldapp.com\/microsoft-fabric-copilot-studio-build-data-agents-over-onelake-analytics\/#breadcrumb"},"inLanguage":"fr-FR","potentialAction":[{"@type":"ReadAction","target":["https:\/\/nboldapp.com\/microsoft-fabric-copilot-studio-build-data-agents-over-onelake-analytics\/"]}]},{"@type":"ImageObject","inLanguage":"fr-FR","@id":"https:\/\/nboldapp.com\/microsoft-fabric-copilot-studio-build-data-agents-over-onelake-analytics\/#primaryimage","url":"https:\/\/nboldapp.com\/wp-content\/uploads\/2025\/09\/image_fd8c8d3c2de46016c6ef2f849a0f73af.jpeg","contentUrl":"https:\/\/nboldapp.com\/wp-content\/uploads\/2025\/09\/image_fd8c8d3c2de46016c6ef2f849a0f73af.jpeg","width":1536,"height":1024,"caption":"Microsoft Fabric + Copilot Studio: Build Data Agents over OneLake Analytics"},{"@type":"BreadcrumbList","@id":"https:\/\/nboldapp.com\/microsoft-fabric-copilot-studio-build-data-agents-over-onelake-analytics\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/nboldapp.com\/"},{"@type":"ListItem","position":2,"name":"Microsoft Fabric + Copilot Studio: Build Data Agents over OneLake Analytics"}]},{"@type":"WebSite","@id":"https:\/\/nboldapp.com\/#website","url":"https:\/\/nboldapp.com\/","name":"nBold","description":"The Collaboration Process Technology.","publisher":{"@id":"https:\/\/nboldapp.com\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/nboldapp.com\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"fr-FR"},{"@type":"Organization","@id":"https:\/\/nboldapp.com\/#organization","name":"nBold","url":"https:\/\/nboldapp.com\/","logo":{"@type":"ImageObject","inLanguage":"fr-FR","@id":"https:\/\/nboldapp.com\/#\/schema\/logo\/image\/","url":"https:\/\/nboldapp.com\/wp-content\/uploads\/2022\/02\/Logo.svg","contentUrl":"https:\/\/nboldapp.com\/wp-content\/uploads\/2022\/02\/Logo.svg","width":74,"height":21,"caption":"nBold"},"image":{"@id":"https:\/\/nboldapp.com\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/x.com\/nboldhq","https:\/\/www.linkedin.com\/company\/nboldhq"]},{"@type":"Person","@id":"https:\/\/nboldapp.com\/#\/schema\/person\/ffd46719c510e6ee95af907cd2c2f985","name":"Alexandre Cipriani","url":"https:\/\/nboldapp.com\/fr\/author\/alexandre-cipriani\/"}]}},"_links":{"self":[{"href":"https:\/\/nboldapp.com\/fr\/wp-json\/wp\/v2\/posts\/13867","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/nboldapp.com\/fr\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/nboldapp.com\/fr\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/nboldapp.com\/fr\/wp-json\/wp\/v2\/users\/11"}],"replies":[{"embeddable":true,"href":"https:\/\/nboldapp.com\/fr\/wp-json\/wp\/v2\/comments?post=13867"}],"version-history":[{"count":0,"href":"https:\/\/nboldapp.com\/fr\/wp-json\/wp\/v2\/posts\/13867\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/nboldapp.com\/fr\/wp-json\/wp\/v2\/media\/13866"}],"wp:attachment":[{"href":"https:\/\/nboldapp.com\/fr\/wp-json\/wp\/v2\/media?parent=13867"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/nboldapp.com\/fr\/wp-json\/wp\/v2\/categories?post=13867"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/nboldapp.com\/fr\/wp-json\/wp\/v2\/tags?post=13867"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}