Abstract—

With a rapidly growing customer base, it has become increasingly difficult for companies to manage and
understand customer data, even with the assistance of Customer Relationship Management (CRM) platforms.

The vast progress achieved in the field of Data Analytics has made it possible to extract meaningful information from raw customer
data efficiently, using which well-informed decisions can be made to enhance customer relationships.

This paper discusses the power of Data Analytics and CRM platforms working in unison, known as CRM Analytics.

It covers the various applications of CRM Analytics and the key data mining techniques used by business organizations. A comparison between top CRM platforms is carried out and a case study on the latest cloud-based Salesforce CRM platform is presented.

 Data Analytics
This is a field that is a narrower perspective of Data Science. Data Analysis is a process of analyzing the data,cleaning it, reconstructing, and modeling data to gain meaningful information, deriving insights, and supporting recommendations

 DATA MINING TECHNIQUES USED IN ANALYTICS
Data mining can be defined as the process of identifying patterns in data from which useful and actionable insights
are gained. There are many data mining techniques that organizations can use, that involve everything from the
basics of data preparation to cutting-edge artificial intelligence. To mention a few key techniques [20][21]:
a) Tracking Patterns
It involves the methodology of noticing trends in the data and monitoring them to make some intelligent inferences.
Once a trend in sales data has been identified, a basis has been established on which actionable insights can be taken .
b) Outlier Detection
It is used to determine aberrations in the data. The anomalies once found by the organization, it can be used to
their advantages as it helps them understand why these outliers exist and occur. It would also help them prepare
and be aware while working on future business outcomes 
c) Association
The data mining technique association is also like the notion of correlation. What this means is that there is a
relationship between the two data events: this could be the simple example of where when someone purchases a cycle
it is frequently accompanied by a purchase of a safety helmet.
d) Classification
Classification is the process of analyzing all the different attributes associated with the kinds of data types that exist.
After which a set of attributes are assigned to each data type and then new data can be easily classified into one of the data types based on its attributes .
e) Regression
This process can be simply defined as a method of identifying the type of relationship that exists between an
output field and every other attribute. It belongs to the category of white-box techniques, there are different types
the most popular being linear and logistic regression.
f) Decision Trees
Decision trees have an extremely forward nature of learning, they too belong to the category of white-box
techniques. They are used to determine how a set of input attributes affect the result or output field. When many
decision trees are combined a random forest is created which is a predictive analytics model

g) Clustering
This data mining technique as the name suggests is used to group similar objects or data into a single entity called a
cluster. Every cluster in independent of every other cluster that is created and this technique is highly used in performing exploratory analysis