ExtendInsights Overview
ExtendInsights lets you schedule automatic data updates to Excel workbooks stored in OneDrive or SharePoint. To achieve this, our app uses the Microsoft Graph API to open and write to your files directly in the cloud. However, due to Microsoft platform limitations and Excel performance constraints, there are important guidelines to follow for reliable scheduling.
🔐 Why Microsoft Graph API Matters
To update a workbook on a schedule, ExtendInsights must use Microsoft’s API to open the file online. This process is not the same as opening the file in Excel desktop — it relies on Excel Online, and thus inherits its constraints.
📏 Microsoft File Size Limitations
Microsoft imposes a 100MB file size limit for opening Excel workbooks in Excel Online (including through APIs). If your file exceeds this limit:
It cannot be opened or written to through the scheduling engine
Scheduled updates via ExtendInsights will fail silently or return an error
This 100MB limit refers only to the current version of the file, not historical versions stored through SharePoint/OneDrive versioning.
⚠️Other Risk Factors Beyond File Size
Even if your file is under 100MB, other factors may still cause performance or compatibility issues:
🧠 Power Pivot Models
Files that include Power Pivot data models often use significantly more memory than the raw file size suggests
These models are evaluated in-memory, and Excel Online may fail to load or write to such files due to resource constraints
🔗 Complex Workbook Logic
Features like Data Models, linked workbooks, volatile formulas, and large PivotTables may degrade performance
Large row counts or high-cardinality columns can also contribute to failures during scheduled writes
✔️ Recommended Best Practice: Use a Source File Approach
To avoid these issues, we strongly recommend the following structure when using ExtendInsights on a schedule:
Create a lightweight source file
Use ExtendInsights to populate this file on a schedule
Keep this file simple — avoid Power Pivot, complex formulas, or visuals
Create a separate analysis workbook
Use Power Query (Get & Transform Data) to connect to the source file
Perform your data transformations, visualizations, and calculations here
This file remains untouched by the scheduler and won’t risk failure
Keeping your source data in one file and your report in another makes sure the scheduler runs smoothly and reliably — while letting you control your data without messing up your analysis.
🛠️ Example Setup
File | SalesData_Source.xlsx | SalesDashboard.xlsx |
Purpose | Scheduled data writes from ExtendInsights | Reporting & analytics via Power Query |
Complexity | Light – no visuals or complex logic | Complex – includes Power Pivot, charts, dashboards |
📝 Summary
It is recommended to keep your source data in a separate file to ensure that Microsoft APIs can open it online during scheduled refreshes without conflicts or delays. This improves reliability and prevents issues caused by file locks or ongoing edits.
We also recommend minimizing the number of stored file versions. Fewer versions help reduce storage usage, improve performance, and lower the risk of the API connecting to an outdated file.
For guidance on managing file versions, see Microsoft’s article:
Enable and configure versioning for a list or library.
❔ Questions
If you need help restructuring your files or optimizing them for scheduled updates, our support team is here to help. Reach out to us via support@cloudextend.io or through the in-app chat.