Dynamics 365 Roadmap: Product Recommendations with Azure Machine Learning
One of the greatest Dynamics 365 feature releases to date is Product Recommendations with Azure Machine Learning. This feature enables Dynamics 365 to use Azure to analyze your current CRM data to automatically predict which products to recommend based on customer purchasing trends.
This predictive and personalized recommendation strategy helps businesses nurture their customer pool, drive faster sales cycles, and thus generate more revenue, especially considering that:
- Increasing customer retention rates by 5% increases profits by 25% to 95%
- Customers report that personal recommendations are the number one reason they repeat business with a company
With this new feature, Microsoft has removed the guesswork when it comes to recommending the right products to the right customers. Previously, CRM administrators were responsible for manually updating product relationships in order to assist the sales team in up-sell/cross-sell efforts. But Azure Machine Learning changes everything.
Currently, Product Recommendations with Azure Machine Learning is available as a public preview. Here, we will walk through what you need to know about this feature and how to enable the preview in your organization.
What is Machine Learning? And specifically, what is Azure Machine Learning?
First, some definitions. Microsoft defines machine learning as “a technique of data science that helps computers learn from existing data in order to forecast future behaviors, outcomes, and trends”. This process involves collecting data from one or multiple sources, and feeding the data into the Machine Learning models. These models then use the data to predict future outcomes. Essentially, Machine Learning uses past data to predict future data.
In regards to Dynamics 365, Azure Machine Learning completely changes the game. Many organizations are currently collecting massive amounts of data on their customer base using Dynamics 365 and have trouble interpreting this pile of information to drive continued product sales.
But connecting CRM data to Azure Cognitive Services allows for the newest Dynamics 365 feature: Product Recommendations. Based on the previous purchasing patterns of your current customer base, Azure Machine Learning can model and predict which products should be recommended to which customers, which ultimately increases the likelihood of a sale.
What does the Product Recommendations preview offer?
Connecting Dynamics 365 to Azure will allow for advanced data modeling capabilities – you will be able to build an advanced recommendation model for automatic cross-sell product recommendations that are based on historical transaction data.
Previously, Dynamics CRM allowed for a system administrator or product manager to define up-sell and cross-sell product relationships. But this method was limited as it relied on static product relationship definitions.
Azure, however, removes the guesswork. It takes what your customers are currently doing to define the up-sell and cross-sell product relationships, and tells your sales team which products to recommend (these are called Suggestions).
Additionally, Azure allows for click-and-drag data analysis. Most organizations that are utilizing advanced data analytics rely on the expertise of data scientists or data researchers. With Azure Machine Learning, your team can set up robust data experiments quickly and effortlessly. Lastly, since Dynamics and Azure are both in the Microsoft family, they are easily integrated to allow for real-time results.
Great! How do I enable the preview for my organization?
In order to set up the Product Recommendations preview, you will need to have system administrator rights to Dynamics 365. Navigate to Settings > Administration > Systems Settings > Previews. Select Cross-sell Product Recommendation Preview, click OK to give your consent, and then click OK to close the System Settings dialog box.
Before proceeding further in CRM, you need to confirm you have an Azure subscription. But wait, I don’t have one! You can set up a free Azure Cognitive Services trial here. After entering some basic information, you can create a new Cognitive Services account within Azure.
The final and critical step is to connect Dynamics 365 to your Azure Cognitive Services account. To do this, you will need the Azure Service URL and Key. You can find the Key under Resource Management when you are looking at the Cognitive Services account.
Once Cognitive Services is set up, return to Dynamics 365 and navigate to Settings > Administration > Azure Recommendation Service Configuration.
This will direct you to set up the Recommendation Connection. Enter the Azure Service URL and Azure Account Key identified above, and then click the Test Connection button. If the test is successful, the Last Connection Status and Last Connection Time fields will be populated and you can click Activate.
Once the connection is Activated, you can customize the Recommendation Models under Settings > Product Catalog > Product Recommendations.
These Product Recommendation Models help define the Product Suggestions. For this example, you can see the Suggestions by opening an Opportunity and clicking the newly added Suggestion Products button in the ribbon.
This article is part of our Dynamics 365 Roadmap series that helps companies stay up to date on the latest Dynamics 365 releases. To see our previous posts, click here to view our roadmap.
As with everything Microsoft, all of this is highly customizable. Hitachi Solutions can help you configure Product Recommendations in your Dynamics 365 instance to match your specific needs. If you want help enabling this preview for your organization, or if you want learn more about Azure Machine Learning, contact Hitachi Solutions today.