Machine Learning for Microsoft Business Central

Add cutting-edge machine learning to your teams tool set

Get a ML Proof-of-Concept

How do you forecast accurately in Business Central?

Machine learning is a fascinating and cutting-edge technology for turning your data into predictive and actionable analytics.

But how do we separate the buzz words from the real value?

Thankfully the tools and technology available today makes these great modeling techniques accessible for organizations of all sizes in a cost-effective way. Our team is able to implement tailored machine learning models directly into your current Business Central & Navision giving your team insights and analysis at their fingertips.

What exactly is Machine Learning?

In attempt to simplify,

Machine learning is a method for training systems to make specific predictions. 

The method involves steps of taking the input data and creating many new representations of that data and then finding the statistical relationships between those representations to make the prediction.

A practical example

Let's say your team is plagued by late payments from customers. Having the ability to predict which invoices will be paid late would prove invaluable in managing cash flow and expenses.

Machine learning is a tool that can be used to predict which invoices will be paid late allowing your team to act proactively, rather than re-actively.

Where do you use Machine learning in Business Central?

Within Business Central & Navision there are standard out-of-the-box methods for enabling Artificial Intelligence & Machine Learning methods. These methods include:

Microsoft Dynamics 365 Business Central Inventory


Predict when you need to replenish inventory.

Microsoft Dynamics 365 Business Central Forecasting


Leverage sales forecasts to generate production plans and create POs.

Microsoft Dynamics 365 Business Central Customer Service

Customer service

Avoid shortages and lost sales by offering substitute items when inventory isn’t available.

Microsoft Dynamics 365 Business Central Financial Planning

Financial planning

Predict how changes to payment terms will impact cash collection cycles.

How do you tailor Machine Learning for your company?

Unique problems requires a unique solution. Although Business Central provides out-of-the-box forecasting and AI enabled functions, the vast majority of problems and insights your team is looking for requires a custom approach.

Each company collects and prioritizes data differently, their problem set is unique, and their way of viewing the market can be vastly different from others. Up until now, to enable machine learning on your companies data would have required your team make a large investment and build up expertise in modeling your data. 

Lucky for us all, times have changed.

How to add custom machine learning models in Business Central & Navision

A deep understanding of the Business Central and Navision data structure and workflow has allowed our team to implement tailored machine learning models in a very cost effective way. This allows clients to get the benefits of the advanced insights without the above mentioned massive time and expense investment.

The underlying data model of Business Central / Navision provides an excellent platform for extension, feature selection, and continuous application of predictive analytics.

The forecasts, predictions, or other outputs from the machine learning models can be viewed on the document level, line level, or in Power BI dashboards and reports linked directly to your Business Central data.

Implementation the ML models

The overarching steps for implementation involve:

  1. Identifying the problem and insights required
  2. A deep dive into the historical data & necessary data clean up
  3. Feature selection & data engineering
  4. Data export & model training
  5. Publishing a call-able web-service for making predictions on new data
  6. Automate predictions and data updates

Step 1-3: Preparing the Business Central data for Machine Learning

Steps 1-3 involve our team understanding what is the problem you're trying to solve. Next we dive into the data you have available and scope out what will be relevant for the machine learning.

To increase the accuracy of the model, a process of feature selection and data engineering is likely required. Our team has a defined blueprint for success that we follow.

Step 4: Data extraction for model training

Step 4 involves moving the groomed data to the Azure Machine Learning Studio where the model is configured, trained, and evaluated. At the end of the training and evaluation stage we will be able to see the accuracy of the model on performing the prediction your team is interested in. At this stage the model is ready to go and your data is removed from Azure as it is no longer needed for the model to operate.

The Azure Machine Learning Studio allows us to easily package your model and publish it as a web service that our team can use to make new predictions.

Step 5-6: Putting the machine learning to work in Business Central

In steps 5-6 our team establishes an automated or semi-automated method for your team to utilize the machine learning in your daily operations or reporting tools.

Using this approach provides a flexible method for integrating cutting-edge machine learning models into your established business processes

see if ml is the right approach

Get a Proof-of-Concept and see if Machine Learning is right for your team.

Before making the investment, our team can perform a proof-of-concept to review if Machine Learning is worth the investment for you.

Review your use cases
Identify the high-value ideas
Review your data
Discuss all ML options and pricing
Provide timeline considerations
Get honest feedback on feasibility
Understand how to get started with ML
Get a POC. Schedule the first call

No obligations, no high-pressure sales & no marketing lists. 
Just real help.