Machine Learning Model can add some additional index to existing data. this data/score is built from learning customer's existing activities/practices and history data maintained in system
While we have machine learning model built on Azure ML, and web service exposed with arranged logic, Further model can be integrated to Customer Insights.
1)Churn Score Machine Learning Score Segmentation- As we add additional churn score to each customer which is some index associated in range of (0-.9). Every customer therefore will have some churn score for example .6, .7 etc.
Segmentation capability can be applied to every attribute associated to customer record, as Churn score is also an addition being added after applying custom ML.
We can project data for All the Customer for which Churn score is greater than .8.this could be your heavily engaging audience.
Configuring segment rule will look like following.
to Create segment from ML, go to create segment from intelligence as shown below
Further configure the conditions-
a) Churn score greater than .8-
b) Spa Usage segmentation
Further segmented data will filter all the records which qualifies the condition.
now this Download csv option allows you to download data/list of customers matching the condition.
2)Recommendation Machine Learning Model Segmentation: Now Another Machine learning custom model could be "Recommendation model". Basically Recommendation model gives first 4-5 recommendation of activity/product based on customers existing history of records configured.
So filter can be based on Customer Who has "Spa" as recommendation, Filter all those customers,
Idea to filter all relevant condition and data to project on, is to get the list of customer we want to target for certain campaign or we might want to create some kind of marketing list to see best suited matched customer for the offerings.
While we have machine learning model built on Azure ML, and web service exposed with arranged logic, Further model can be integrated to Customer Insights.
1)Churn Score Machine Learning Score Segmentation- As we add additional churn score to each customer which is some index associated in range of (0-.9). Every customer therefore will have some churn score for example .6, .7 etc.
Segmentation capability can be applied to every attribute associated to customer record, as Churn score is also an addition being added after applying custom ML.
We can project data for All the Customer for which Churn score is greater than .8.this could be your heavily engaging audience.
Configuring segment rule will look like following.
to Create segment from ML, go to create segment from intelligence as shown below
Further configure the conditions-
a) Churn score greater than .8-
Further segmented data will filter all the records which qualifies the condition.
now this Download csv option allows you to download data/list of customers matching the condition.
2)Recommendation Machine Learning Model Segmentation: Now Another Machine learning custom model could be "Recommendation model". Basically Recommendation model gives first 4-5 recommendation of activity/product based on customers existing history of records configured.
So filter can be based on Customer Who has "Spa" as recommendation, Filter all those customers,
Idea to filter all relevant condition and data to project on, is to get the list of customer we want to target for certain campaign or we might want to create some kind of marketing list to see best suited matched customer for the offerings.
