Can you imagine how a company can be impacted by a system which will suggest products on the basis of past data?
This can help the Sales team target products better to their existing clientele and in the process have better outcomes and generate more revenue for the company.
Dynamics 365 comes with a new feature called Product recommendations with Azure Machine Learning. Currently, it is still in the preview mode but offers the customers lots of benefits when it comes to product recommendation. Product Recommendation from Azure Learning employs Machine Learning, so before going into the configuration of the product recommendation feature on CRM, let’s have a look at what is Machine Learning.
Machine Learning is a method of data analysis that automates analytical model building. Using algorithms that iteratively learn from data, machine learning allows computers to find hidden insights without being explicitly programmed where to look. Computers apply different learning techniques to automatically identify patterns in data and these techniques can be used to make highly accurate predictions.
So how does Dynamics 365 recommend products when you create opportunities?
Dynamics 365 is integrated with Azure Cognitive Service APIs, using which it gives the recommendation of products.
What is Azure Cognitive Services?
It is a collection of APIs, that enable natural and contextual interaction with tools that augment user experiences via the power of machine learnt models from Microsoft. Some APIs available are Recommendations, Emotion, Face Recognition, Language, Speech and Text Analytics.
You can find an excellent video over here which talks about Microsoft Cognitive Services.
To enable cross-sell product recommendations feature follow the below steps
Before creating the models we need to configure the connections to Azure cognitive service API.
This url is different for each region.
Create Recommendation Model
Firstly, we need to create a model for automatic cross-sell product recommendations based on historical transaction data. This data will be passed to Cognitive Recommendation service for predicting the cross-sell products. Follow the below steps after configuring the product recommendation connection
3. Click on the Build Model Version. On clicking of the Build Model Version a window popups with a default name
4. Click OK
5. We can see the new Azure Model Id and Build version shown in the screen
6. Wait for couple minutes as it has to synchronize all the opportunity/quote/order transaction details to Azure.
7. Once the Model is built it will look like this. Notice the “Azure Model Build Status”, “Catalog Synchronization Status” and “Basket Data Synchronization Status”.
8. Now let’s test the recommendation. Click on the Test Recommendations button, it will display a popup and enter the details.
9. We can see the Product Recommendations in the list. Click on Close.
10. Now set Recommendation Version field value to the version we just generated.
11. Click on Activate.
Testing the Recommendation in real time
This is OOTB Dynamics 365 integration with Azure Cognitive service. There are a plethora of things which can be achieved using Azure Machine Learning and Cognitive Services. The possibilities are endless.
Ever changing hobbies ...but an undying passion for Dynamics CRM, Salesforce and all the new age technologies out there!
Shipping and Transportation provider reaps benefits from a CRM overhaul
Customized CRM for one of the fastest growing mortgage companies
Dynamics CRM integrated with SharePoint to deliver a simplified sales and marketing platform for a manufacturing giant