Difference between revisions of "AY1516 T2 Team13 Natasha Studio Project Overview Methodology"

From Analytics Practicum
Jump to navigation Jump to search
Line 40: Line 40:
 
==<div style="background: #B0C4DE; padding: 10px; font-family:Arial; font-size: 14px; font-weight: bold; line-height: 1em; text-indent: 3px; padding 0px 18px 0 px 18 px;"><font color="white">SCOPE OF WORK</font></div>==
 
==<div style="background: #B0C4DE; padding: 10px; font-family:Arial; font-size: 14px; font-weight: bold; line-height: 1em; text-indent: 3px; padding 0px 18px 0 px 18 px;"><font color="white">SCOPE OF WORK</font></div>==
  
<b><u>Proposed Techniques</u></b>
+
[[File:scope.png|800px|center]]
<br>1. Association Rule Learning (Affinity analysis) will be performed on the purchases of dance packages to further understand the purchase behavior of customers. The results will enable us to implement suitable sales promotions, cross-selling promotions and recommendations in order to capture more sales per customer.
 
<br><br>2. Event Sequence Analysis will be used to analyze the purchasing patterns of customers. This allows us to understand the trends and decisions made by the customers after the expiration of every package. Following which, suitable sales and marketing efforts can be implemented in order to drive repeat sales.
 
<br><br>3. Time Series Analysis will be used in conjunction with other analytical methods to determine if there are correlations between the customers’ purchasing patterns and behavior. For example, the amount of customers who have repeat purchases after wastage of any monthly packages.
 
<br><br>4. Analytical Customer Relationship Management techniques will be used to interpret and report customer-related data to enhance both customer and company value. For example, customers will be segmented based on a customer scoring mechanism for lifetime value. The scoring mechanism will be based on certain factors such as “Frequency of purchase” and “Total life time purchase”. After which, personalization marketing can be applied to the relevant segments.
 
  
 
==<div style="background: #B0C4DE; padding: 10px; font-family:Arial; font-size: 14px; font-weight: bold; line-height: 1em; text-indent: 3px; padding 0px 18px 0 px 18 px;"><font color="white">METHODOLOGY</font></div>==
 
==<div style="background: #B0C4DE; padding: 10px; font-family:Arial; font-size: 14px; font-weight: bold; line-height: 1em; text-indent: 3px; padding 0px 18px 0 px 18 px;"><font color="white">METHODOLOGY</font></div>==

Revision as of 19:25, 28 February 2016

HOME

TEAM

PROJECT OVERVIEW

FINDINGS & ANALYSIS

PROJECT MANAGEMENT

DOCUMENTATION

BACKGROUND DATA METHODOLOGY


SCOPE OF WORK

Scope.png

METHODOLOGY

Tools used
The main tools that will be used are Microsoft Excel and R. Microsoft Excel will be used for the data preparation and cleansing process as it is preferred by our sponsor. R will be used mainly to build and evaluate the models. The open source nature of R would also allow our client to also use it with future data.

Data Extraction
At present, data from July 2010 – August 2012 is available to us on Microsoft Excel. At Fall 2012, the business owner decided to stop using the system. Thus, purchases and attendance were recorded on paper. As a result, work has to be done during the data assessment process to enter data into the spreadsheet so that there are more data to work with.

Data Preparation
In addition to the data that is missing from 2013-2015, the current data set that was presented to us would require significant efforts in data cleansing due to the inconsistency, missing data fields and duplicates that has been resulted from bad practices throughout the 3 years. We foresee that huge amount of time will be used for this process.

Data Validity
As of the data that was presented to us, there were a total of 1717 members and 3044 purchases. The numbers are not final as data cleansing has yet to been done during this stage of the project. However, based on visual observation of the data, we are confident that it should not deviate too far away from the reported numbers as shown above. In order to perform a substantial analysis on the data, a rough estimate of 5000 data points is required for the proposed techniques shown below. The current data set is not sufficient as it does not meet the required sample size. However, the business owner has informed us that he has a rough estimate of 5300 members. As such, we are confident that after the data assessment process, we will have sufficient data to work on.