Open the data model

[b-kopik]

Hello,

I just started my new role and I need to update the dashboards in Power BI Apps. I tried opening the data model for by selecting More options (...) and Open data model. However, I am getting an error message. Is there a way to fix that?

I could use Power BI dashboard but then I would have to update all the links.

Thanks,

Bruna

[v-tsaipranay · Accepted solution]

Hi @b-kopik ,

We haven’t received an update from you in some time. Could you please let us know if the issue has been resolved?
If you still require support, please let us know, we are happy to assist you.  

Thank you.

[b-kopik]

Thanks for all the help!

[v-tsaipranay]

Hi @b-kopik ,

I wanted to check if you had the opportunity to review the information provided. Please feel free to contact us if you have any further questions.

Thank you.

[v-tsaipranay]

Hello @b-kopik ,

Just to clarify, storing data in a Fabric Lakehouse is different from SharePoint. In SharePoint, you can still download the original Excel file, but in Lakehouse, the file is ingested into OneLake and stored as structured data (Delta/Parquet format) for analytics. So, users can't download the original Excel file directly from Lakehouse like they can from SharePoint. If it's important to have the raw Excel files available for download, SharePoint is better for file sharing, while Lakehouse is best for analytics and scalable storage.

About the file location in Lakehouse: if you only see the file name in the table view, the data has been ingested into a table and the original Excel file isn't stored as-is. To find physical files, go to your Lakehouse and check the Files section where uploaded files like Excel, CSV, and Parquet will show up there. Tables created from these files are stored separately in the Tables section.

This approach lets you keep Excel files in SharePoint for downloads, while using Lakehouse for long-term analytics and governance.

Thank you.

[b-kopik]

Thank you. I am wondering if we need Lakehouse then, given the size of our datasets and since we would like to have the ability to download the data as needed.

[v-tsaipranay]

Hi @b-kopik ,

Based on your current scenario with a few small files under 10,000 rows and the need for managers to download Excel files, sticking with SharePoint and Power BI is the most practical approach. This setup is straightforward and allows users to access the original files easily.

Lakehouse solutions are generally more beneficial when you have larger data volumes, need to integrate multiple sources, or require advanced governance and performance features. For now, SharePoint is the best choice, and you can consider moving to a Lakehouse setup in the future if your needs change.

Hope this helps, Please feel free to reach any further assistance.

Thank you.

[v-tsaipranay]

Hi @b-kopik ,

I wanted to follow up on our previous suggestions. We would like to hear back from you to ensure we can assist you further.

Thank you.

[v-tsaipranay]

Hi @b-kopik ,

Thank you for your follow-up questions. To clarify, taking over the dataset will not affect the existing modelling work tables, relationships, and measures will all remain unchanged. The only difference is that the previous credentials are removed, so you’ll need to enter your own SharePoint credentials for the dataset to refresh. If you see multiple SharePoint items, it’s likely because several reports or dataflows are connected to the same SharePoint source. To identify the exact file(s) linked to your dashboard, go to your semantic model settings and select Gateway and cloud connections. There, you can view the full SharePoint file path(s) associated with the dataset and match them with your SharePoint folders. After updating your credentials in this section, your dataset will connect successfully without altering the existing model logic.

Thank you.

[b-kopik]

Hello,

Thank you so much for looking into this. I spent the last week trying to understand the model but the current setup is complex, prone to errors, contains lots of old data, causing slow updates and difficult maintenance. Also, all managers want to make updates to the dashboards.

Since my team is only using 4 relatively simple dashboards (each using about 2 data files with no more than 10,000 rows of data each), I think I will do the following:

I’ll keep things simple and editable in Power BI for now while managers provide feedback and we finalize the updates. Once everything is locked in, I’ll transition the cleaned data to Fabric's Lakehouse for better structure, performance, and long-term scalability.

Any thoughts on this approach? I would appreaciate any recommendations.

Thanks again.

[b-kopik]

Hi,

Thank you so much! That makes sense. Will users be able to download the Excel files from Lakehouse though? I think the preference would be to use Sharepoint for that reason. What I am struggling now is that I can see the file name on Lakehouse but I can't locate it anywhere.

Thanks again.

Bruna

[v-tsaipranay]

Hi @b-kopik ,

Thank you for sharing information about your current setup and proposed approach. Your plan seems practical and keeping the models simple and editable in Power BI at this stage will make it easier for managers to give feedback and for you to make adjustments. When the design is stable, moving the cleaned data to Fabric’s Lakehouse is a solid option for scalability and performance, as it offers a centralized and governed data layer that can efficiently support multiple dashboards.

As additional recommendations, as you move to the Lakehouse, it’s a good idea to establish clear naming conventions and folder structures early so your team and managers can easily navigate and maintain the data.

For performance, when your datasets become larger or more complex, enabling incremental refresh in Power BI will help improve refresh times. Since multiple managers require editing capability, you may also want to use workspaces with clearly defined roles (Viewer, Contributor, Member) to manage permissions and prevent unintentional changes. Looking ahead, Fabric Lakehouse will give you the flexibility to scale beyond SharePoint files and integrate with other enterprise data sources if needed.

Overall, your approach of starting simple and then moving into Fabric for structure and performance later is aligned with best practices.

Hope this helps, Please feel free to reach any further assistance.

Thank you.

[v-tsaipranay]

Hi @b-kopik ,

Thank you for the update, and I’m glad you were able to view the model through the Lineage view. In Power BI Service, if you’re not seeing a separate “dataset” object, it usually means the report is either connected directly to an external data source (in your case, SharePoint files) or the dataset is stored in a different workspace.

In this case, the semantic model you see in the workspace is the dataset. You can use the Lineage view to identify the connected SharePoint files and verify their locations:

1. In the workspace, open the Lineage view (icon shown earlier).
2. Select the semantic model connected to your report.
3. In the right-hand pane, look at the Data Source settings or Details section this will list the SharePoint file paths or sources.
4. Match these paths with your SharePoint access to locate the exact files feeding the dashboard.

If you don’t see the file paths, you may still need Build permission on the semantic model. Please request this from the dataset owner so you can inspect the connection details.

Here’s an example of how the Lineage view will display the connection to a SharePoint or Excel file:

Once you identify the files, you can manage refresh settings and updates directly, ensuring your dashboards remain up to date.

Hope this helps. Please reach out for further assistance.

Thank you.

[b-kopik]

Hello,

Thank you so much for the background. I tried to follow your steps as the screenshot is exactly what I am looking for, but I keep getting this error message.

Do I have to take over? This individual has left the organization:

Thanks,

Bruna

[v-tsaipranay]

Hi @b-kopik ,

Thank you for providing the details and screenshots; this helps clarify the situation. The error occurs because the dataset was set up by a previous employee, and their credentials are no longer valid.

To resolve this, you will need to Take over the dataset. Doing so will:

• Make you the new owner of the dataset and remove the previous user's credentials.
• Let you enter your own SharePoint credentials under Gateway and cloud connections so the refresh can work again.

Once you take over, remember to update the dataset settings, including credentials for each data source, and set up any scheduled refreshes again.

Afterward, try refreshing the dataset; it should connect successfully with your credentials.

I hope this helps clarify things. Please feel free to reach out if you have any further questions.

Thank you.

[b-kopik]

Thanks! If I take over, will it override any edits that the previous user did to the model? I don't want to make any changes to the modelling etc.

[b-kopik]

I also don't understand how I can find out the exact data source since there are multiple under the same name and I can see is this:

[v-tsaipranay]

Hi @b-kopik ,

Thank you for reaching out to the Microsoft fabric community forum and I appreciate the helpful guidance already provided by @rohit1991 .

They are correct “Open data model” requires Build permission on the dataset or Contributor/Member access to the workspace where the dataset resides.

Since your reports connect to SharePoint files and you’re unsure which files belong to each dashboard, I recommend following @Idrissshatila 's  suggestion to open the Lineage view in your workspace. This will help you trace the semantic model back to its source and identify the connected SharePoint files. Once you know the exact file locations, you can verify your access and refresh settings to manage the data as needed.

Hope this helps. Please reach out for further assistance.

Thank you.

[b-kopik]

Thank you! They have enabled Fabric for me. They said they will assign admin role for me so I can access the premium workspaces  the previous analyst used. Do you think that will work?

I only see the report and semantic model in the workspace - no datasets. The data is linked to files in sharepoint. Does that mean I won't be able to refresh the data in PowerBI?

[rohit1991]

Hi @b-kopik 

Even with Fabric enabled and admin rights in the premium workspace, you still won’t be able to refresh data unless the dataset itself exists in that workspace.

If you’re only seeing the report and semantic model but no dataset, that usually means the report is connected live to another dataset (possibly in a different workspace) or is pointing directly to SharePoint/Excel files.

• If it’s a live connection to another dataset, you’ll need build permission on that source dataset to refresh or modify it.

• If it’s connecting directly to SharePoint files, the refresh will depend on whether the connection settings and credentials are configured in the service.

So, admin rights in the current workspace alone won’t solve it ,you also need access to the actual dataset or data source the report is using.

[b-kopik]

Thank you so much. It looks like the data is connected to SharePoint files, which I have access to. However, I don't know which files belong to each dashboard. Is there a way to find out? Since I can't access the data model, I can't look into the data source settings and figure out what dataset it is linked to.