Visualization of Consumer Satisfaction Project Overview

From Analytics Practicum
Revision as of 17:14, 27 September 2014 by Mohamedsh.2011 (talk | contribs)
Jump to navigation Jump to search
Home   Project Overview   Documentation


Background

Consumer research has been a hot topic. Businesses and government agencies are interested to know the satisfaction levels of Singaporean consumers and effectively take actions that can create valuable and meaningful impact in the society. This project explores these satisfaction levels. It uses the respondent level data from the Customer Satisfaction Index of Singapore (2008-2013) for the following sub-sectors:

  • MRT System
  • Public Buses
  • Taxi Services
  • Supermarkets
  • Hotels
  • Budget Airlines
  • The metrics that measure satisfaction includes:
    Csisg.jpg
    Note: Referenced from the CSISG Brochure

    Motivation

    The motivation for doing this project was based on our interest in presenting the data in an interesting way that will help users and business leaders derive fresh insights from the data. Through the visualizations, we hope to illustrate customer satisfaction in various measures as shown in the CSISG Causal Model above to provide the users with a better understanding of the data.

    Objective

    The objective of this project is to produce an interactive dashboard that shows market trends and customer satisfaction visually. We will try to incorporate a ‘fun’ element into the dashboard so that users and business leaders will be interested to play with it during events and roadshows organized by the Institute of Service Excellence.

    Approach

    We will be developing the dashboard using the d3.js library primarily. During the initial stages, we will be data analysis tool SAS Enterprise guide to conduct Exploratory Data Analysis to drill down on the more important information to be shown in the dashboard. Next, we will use data visualization tools such as Tableau and to plot the data in different charts and decide on which charts are more suitable to present the data in. We will then proceed to come up with prototypes of our dashboard and consult our project sponsor and supervisor before continuing with the development of our dashboard finally.


    Project Risks

    Risk & Mitigation Plan (From Acceptance)

    <br

    Types of Risk Risk Description Likelihood Impact Level Mitigation
    Project Management Risk

    Delay in completion of task. These task can cause possible bottlenecks due to dependency requirements. This may affect the delivery of the product.

    High

    Medium

    A

    Includes buffer in every iterations in order to deal with such situation. Do a clearly defined critical path to ensure a clear understanding on the critical path. More time and effort should be given core components that affects the success rate of other components. Understand which task can be brought forward in the event other task cannot be completed without prior pre-requisite component.

    Client may potentially have different ideas and suggestion along the way. This may translate into a scope creep , especially if the client were to venture into areas not discussed previously.

    Medium

    High

    A

    Change management Plan. Review each change request case by case.Discuss and interact with the client on the feasibility and priority of change. If a change is necessary, re-confirm priority of function with client again. A Change request may be rejected if the team is lacking resources such as time or man-power.

    Technical Risk

    Team is unfamiliar with the current technologies used such D3. All of the core components requires the use of D3 to interact with external sources like CSV files . There may be a need to involve other technologies and this could lead to integrating issues.

    High

    High

    A

    Study these technologies off-hand to be more prepare when development portion began. Technologies must be well understood for implementation, hence the Visual Analysts should explore the client's existing codes in order to utilize and re-usable codes. This also allows the Visual Analyst to understand the current logic that was implemented.


    High volumes of data is stored by the client. Potentially these data may not be sufficiently clean and suitable for analysis. Most data derived from database management systems is geared towards speedy performance and transaction and not for Business Analytics.

    Medium

    Medium

    B

    Create new database structure specifically for business analytic. This is to ensure that the current database can still be used for fast performance while the new database structure free for study. Additional time will be given to extract, load and transform data to the most appropriate forms.

    The project is implement a newly proposed dashboard. This is a problem as there is no previous reference for both the client and visual analyst.

    Medium

    Medium

    B

    During the exploration phase, the team can write comments base on what he/she has discover about a particular design. This helps those who have yet learn that part of the design to better follow the reasoning . Interaction with client frequently gives us a better understanding on the requirement. Also, consulting VA Prof may help.

    Difficulty in integrating multiple systems together like Web service, database server and Visual analytics since the team does not have anything to work with. Team needs to pass the codes to the client so that he can test and check that it integrates seamlessly.

    Low

    High

    B

    Time will be taken to understand the limitation of the technologies used




    Risk & Mitigation Plan (From Mid-Term)

    This Risk Table Structure was inspired by ThunderBolt

    Types of Risk Risk Description Likelihood Impact Level Mitigation
    Project Management Risk

    Team members might underestimate school commitments and conflicting priorities. This might affect the completion of deliverables and even the quality.

    Medium

    Medium

    B

    Team members will meet up and discuss to clarify if there are other commitments they need to adhere to. PM will revise the schedule by including sufficient buffer days during such "busy" periods.


    Team might not be able to complete all the agreed upon features by final presentation.

    Low

    High

    B

    Team will liaise with client and make client understand clearly why there might be a potential drop in functions i.e. due to lack of time according to schedule

    Technical Risk

    During the official tournaments, a critical bug may be found like users might not be able to join a group tournament on SingPath.

    Medium

    High

    A

    Before the tournaments, the team will hold a mock tournament within SMU to source out critical bugs and solve them. This would be our 2nd UT.