Marketing Analytics Week3
Methods of Marketing
Analytics
Marketing Analytical Methods
 Descriptive Analytics
 Diagnostic Analytics
 Predictive Analytics
 Prescriptive Analytics
Descriptive Marketing Analysis Transforming data into useful insights
 The process of finding useful and important information by analysing the huge data
 What happened in the past? By using the stored data.
 Data stored in data warehouses. Limited to past data.
 ETL process is used before Warehousing of data
 It provides accurate data in the reports using past data.
 It allows the reactive approach
 Sales report, revenue of a company, Traffic Growth of Telco
 Examples : Reports, Dashboards, dataViz, Graphs
Diagnostic Marketing Analysis
 Takes the insights found from descriptive analytics
 Drills down to find the causes of those outcomes
 Creating connections between data and identifies patterns of behaviour
 Examples : Trends deviation, Root Cause, Comparative Data Analysis
.
Diagnostic Marketing Analysis – Real
World Example – Telecom Industry
 Increase in Network Congestion after Rainfall
 Hardware malfunctioning after rainfall causes Availability issues on some sites
 Load increased on Neighbors and congestion increased
 RCA shows 3G major culprit cells, Action taken & issue resolved
Average of Cell Availability Rate(%)
3G Site 10/30/2020 11:00 10/31/2020 11:00 11/1/2020 11:00
RYK01004A
0.00%
0.00%
0.00%
RYK01004B
0.00%
0.00%
0.00%
RUR20353A
0.00%
0.00%
0.00%
RUR20353B
0.00%
0.00%
0.00%
RYK72722A
0.00%
0.00%
0.00%
RYK72722B
0.00%
0.00%
0.00%
Predictive Marketing Analysis Transforming data into future insights
 Predictive analytics use models constructed from past data to forecast future
outcomes based on historical data.
 Involves Statistics and forecast techniques
 Results are not 100% accurate
 it will not tell you exactly what will happen but it will tell you what might happen in
the future.
 This a proactive approach
Predictive Marketing Analysis
Techniques
 Classification
 Regression
 Clustering
 Time Series
Classification Model
 These models work by classifying information based on historical data
 Classification models are used in different industries because they can be easily
retrained with new data and can provide a broad analysis for answering questions
Regression
 Estimate Relationship between Variables
 Predict Value of dependent variable on the basis of independent variable(s)
 Simple regression involves a single independent variable.
 A single straight line is fit to the data
Y = b0 + b1X
where
b0 is the intercept
b1 is the slope
 Multiple regression involves has more than one independent variable
Time series model
 Time series model focuses on data where time is the input parameter
 If organisations want to see how a particular variable changes over time, then they
need a Time Series predictive analytics model
 it can take into account extraneous factors that could affect the variables, like
seasons.
 Telecom Packages Pricing Based on Utilization
Clustering model
 sorts it into different groups based on common attributes
 The ability to divide data into different datasets based on specific attributes is
particularly useful in certain applications, like marketing
 For example, marketers can divide a potential customer base based on common
attributes
 Another example, Create a pilot cluster for trial and roll out changes if trail has
successful results
Predictive analytics uses
 Segment customers
 Reduce customer churn
 Retain profitable customers
 Develop effective campaigns
Prescriptive Analytics - What do I need to do?
 The prescriptive model utilises an understanding of what has happened, why it has
happened and a variety of “what-might-happen” analysis to help the user determine the
best course of action to take
 Examples : Google Ads, Facebook Ads
THANK YOU