ANLY482 AY2016-17 T2 Group21 : PROJECT OVERVIEW

PROJECTS

HOME

ABOUT US

PROJECT OVERVIEW

PROJECT FINDINGS

PROJECT MANAGEMENT

DOCUMENTATION

Project Overview

Dressabelle is a leading online fast-fashion retailer in Singapore. Having established themselves online since 2012, Dressabelle have also expanded to 6 physical stores in malls around Singapore. Dressabelle set themselves apart with not one but two fresh fashion collection every week. Their mission is for women to feel confident about themselves when they dress in their clothes. With affordable pricing they want women to know they don’t have to blow a couple hundred for a dress that doesn’t compromise on style or quality.

Business Problems

Being in the fast-fashion industry, Dressabelle launches new fashion collection each week, the fast-paced nature of their merchandising, logistics and marketing has given little room to explore how data could aid their decision process.

For merchandising, know what to purchase, how much to purchase are important business decisions that have to be made on a regular basis. Dressabelle lacks a data-driven feedback loop on which products have done well and which can be done better.

Part and parcel of marketing is to attract new and old customers, dressabelle faces two main challenges. First, they are unclear how well are their marketing efforts are paying off, current marketing investments into Google and Facebook do not have clear return of investments. Second, it is crucial to identify which customers to target their marketing efforts into, current practice includes promotions to 'lost' customers.

Motivation

Dressabelle aims to be a leading player in the fashion industry, this require them to kept in the know of the latest fashion trends, streamline their internal operations and as well as understand their customers. The use of analytics for merchandising and marketing will give them a competitive advantage over other players in the market.

Project Objectives

The aim of this project is to help 4FINGERS understand and improve sales on 3 levels:

1. Product-level analytics - Analysing purchasing patterns and identifying commonly brought product combinations to find up-selling and promotional opportunities

2. Customer-segment level analytics - Identifying distinct groups of customers such as those who dine-in, take-away or call for delivery. Customers coming at different times of the day can exhibit different purchasing behavior

3. Outlet-level analytics - Performing analysis on outlet-level help identify location specific trends and demand. Testing and learning from our recommendations starting from a single outlet will help validate our hypothesis.

Project Scope

Data collection - Gather data from 4FINGERS POS systems and product catalog

Data preparation - Cleaning data and anonymising/censoring data

Analysis of Market Baskets - Generation of association rules and contextualising it with product information.

Analysis of Customers segments - Grouping market Basket analysis by customer segments data

Analysis of Outlet trends - Grouping market Basket analysis by outlet level data

Refinement - Get client feedback and refine our models

Stakeholders

Project Supervisor: Prof Kam Tin Seong, Associate Professor of Information Systems; Senior Advisor, SIS

Sponsor: Gwen Ang, Marketing Manager, Dressabelle