Sales and Inventory Forecast in Dynamics 365 Business Central

Manual forecasting often leads to costly mistakes and missed opportunities. The Sales and Inventory Forecast feature in Dynamics 365 Business Central leverages Microsoft Azure AI to help businesses predict future demand and plan replenishment more accurately, helping you move from reactive ordering to a proactive strategy.

At Encore, we’ve helped businesses unlock this feature to reduce risk, improve working capital, and make confident supply chain decisions. In this blog, I’ll walk through how you can set it up and start using the Sales and Inventory Forecast feature today.

Table of Contents:

• What Is Sales and Inventory Forecasting
• Sales and Inventory Forecast Setup
• Setting Up the Job Queue
• Forecasting Assistant: Copilot
• Troubleshooting
  • Forecast Not Generating
  • Job Queue Errors
  • Forecast Doesn’t Reflect Open Purchase Orders

What Is Sales and Inventory Forecasting in Business Central?

This Business Central extension leverages AI-driven insights based on historical demand, supply, and inventory data. With more accurate predictions, you can use it to:

• Reduce stockouts
• Minimize over-purchasing
• Maintain healthier working capital

While it was once a manual add-on, Microsoft has made this extension default and globally published in recent versions of Business Central; it is now a native, pre-installed, active extension.

Sales and Inventory Forecast Setup

Where to find the extension:

You can verify the extension is active by navigating to the Extension Management page; it is titled “Sales and Inventory Forecast” by Microsoft.

Option 1: Using Tell Me (Search)

1. Go to Tell Me (Alt + Q).
2. Search for “Sales and Inventory Forecast Setup”.
3. Open the page.

Option 2: From an Item Card

Open any Item Card, then scroll to the Sales & Inventory Forecast FactBox on the right-hand side of the screen.

Sales and Inventory Forecast Parameters

On the Sales and Inventory Forecast Setup page, the user can control key parameters:

Period Type, Horizon, and Stock Warning Horizon

Use these parameters to define how far into the future forecasts cover and when items are expected to run out of inventory. In the example above, a Stockout warning occurs for up to 3 months out, and the total time covered would be 12 months out.

Variance %

Under Variance %, the user can control for the deviation allowed, plus or minus. In this example, any variances exceeding 40% would result in the application not making a forecast. In this case, the system on the Item Page FactBox will generate an error signaling a status of “Variance too high”:

Note: Forecasts will not be generated if there is a lack of sufficient prior data. This data is based on item ledger entries, for sale type transactions with items involved.

Expiration Period

The Expiration Period can be set to determine the number of days until the forecast disappears, after which new forecasts will be generated.

Historical Periods

Historical Periods indicate how many past periods of data are analyzed for future forecasts.

Timeseries Model

And finally, the Timeseries Model specifies the algorithm used for the time series analysis. More information on these models can be found in the Machine Learning Mastery article: 11 Classical Time Series Forecasting Methods in Python (Cheat Sheet).

Azure AI Integration: What Changed

Although this is still technically a Dynamics 365 Business Central extension, Microsoft now manages the Azure AI connection by default. You no longer need to manually configure API keys unless you have advanced custom needs.

In older versions, you could see the “Open Azure AI Gallery” on the setup page. This allowed users to clone an experiment from the Azure AI Gallery and manually enter the API URI and API Key into the Sales and Inventory Forecast Setup parameters.

In recent versions, Microsoft has simplified the process. The connection to Azure AI is now pre-configured and managed by Microsoft.

• The Change: Microsoft no longer expects users to manually “clone” experiments unless they have highly advanced custom needs.
• How to check: If the API URL and API Key fields are blank, Business Central is using the default Microsoft-managed service. You don’t need the Gallery link; the forecast will work automatically as long as you have enough historical data.

Set Up the Job Queue

Once your Sales and Inventory Forecast Setup parameters are configured on the setup page, we can click either “Setup Schedule Forecasting” to set up the job queue or “Update Forecast” to manually trigger the update of the forecast data.

In the Scheduled Forecasting Job Queue, with status set to “On Hold”, the user can set up parameters to run the sales forecast:

In the above setup, the forecast is run daily overnight, with 3 attempts before stopping should an error arise. When the setup is complete, change the status from ‘On Hold’ to ‘Ready’ via the Process tab.

You can also update a specific item’s Forecast on demand by clicking on Actions > Forecast > Update Sales Forecast on the Item ard.

On Item Cards, the forecast will display as a predictive graph. From there, you can create purchase documents with quantities pre-filled based on the missing quantity to prevent stockout inventory:

If you create a purchase invoice and purchase order but do not invoice it and leave the document open and outstanding, the system will not account for those inventory quantities. Therefore, if you choose to create a purchase invoice or purchase order before posting the prior entries for the item, the system will generate the same quantities for the item. This would also apply to sales documents in the same manner.

Your Forecasting Assistant: Copilot

Dynamics 365 Business Central’s integrated Copilot enhances forecasting by helping users understand patterns, ask questions, and make decisions more quickly.

How Copilot Helps

• Explains forecast trends in natural language
• Suggests replenishment quantities
• Identifies unusual spikes or drops
• Highlights items at risk of stockout
• Converts forecasting insights directly into purchase orders

For example, you can ask Copilot:

“Explain why Item 1000 has a steep demand increase next month.”

Copilot replies with sales trend reasoning and external factors (seasonal patterns, historical spikes).

Watch our recorded webinar for a full walkthrough of Sales and Inventory Forecasting with Copilot in Dynamics 365 Business Central:

Common Issues and Troubleshooting

Forecast Not Generated

If any issue occurs that prevents forecast data from being generated, the FactBox will display it in the status file. By clicking on the status, you can gain a better understanding of the root causes of the issue.

In this case, it is “The calculated forecast shows a degree of variance that is higher than the setup allows.”

Possible reasons:

• Not enough historical sales
• High variance
• Incorrect Period Type
• The item is blocked

Job Queue Errors

If the job queue fails, check:

• User permissions (for the user account running the Job Queue Entry)
• Environment update in progress
• Time conflicts with other jobs
• Missing Azure service connection (rare)

Typically, if an error occurs involving the Job Queue, it is recommended to contact your administrator for support in resolving the issue.

Forecast Doesn’t Reflect Open Purchase Orders

Business Central does not include unposted purchase or sales documents in forecasts; only posted Item Ledger Entries are counted.

---

As a Dynamics 365 Business Central Partner, Encore helps clients leverage advanced features in Dynamics 365 solutions. We know that training and support are essential for your team to get the most out of Business Central.

For more walkthroughs of functionality in Business Central, you may be interested in these articles:

• Analysis Mode in Dynamics 365 Business Central
• Impactful Copilot Features in Dynamics 365 Business Central