Big W - QBUS6600 Project Outline
Background
BIG W (https://www.bigw.com.au/) is part of the Woolworths Group. Operating nearly 180
stores nationwide and employing over 18,000 people, it is a discount department store focused
on selling general merchandise and everyday goods.
The discount department store market is highly competitive; BIG W’s major traditional
competitors are Kmart and Target. BIG W also competes with the supermarket business unit
of the Woolworths Group and the Coles supermarket chain.
BIG W’s core customers as a discount department store tend to be price sensitive and will
often cross-shop with competitors in an effort to find the best offer for the products they are
buying.
In recent years, BIG W’s customers have been struggling due to the prevailing economic
conditions, with the cost of living increasing and the effects of inflation. They will typically have
less disposable income and less purchasing power than in previous years and, therefore, will
be highly discerning about where to spend their money. This has resulted in a tougher trading
environment for BIG W. BIG W’s stated purpose is to “make a real difference for families” - to
this end, Woolworths Group aims to find a way to sell more products to BIG W customers
whilst maintaining their gross profit. This goal will allow the business to deliver more value to
their customers without damaging their profitability, which is of particular concern in the
aforementioned difficult trading conditions in Australia.
Mechanics of Pricing, Promotions and Availability
Pricing
BIG W has a national pricing model, which means that the selling price for an article is the
same in every store.
• The following limits are applied to the frequency and amount of price changes to satisfy
legal requirements and limit the burden on stores for changing product tickets:
o The maximum number of total price changes (total number of products with a
price change applied), excluding promotional price changes for a single week,
is 20.
o Price changes are applied on a Tuesday and should be assumed to take effect
for all transactions onwards including those on the Tuesday, until such time as
another price change is applied. Any price change must remain in effect for a
minimum of six weeks.
• The provided competitor pricing data is not exhaustive. Changes in price should take
this into account.
Promotions
BIG W runs “promotions” for specifically identified articles (products) for limited durations in
partnership with the vendors supplying the product. Note the following on promotions:
• BIG W runs promotions (discounted pricing) on a regular basis. These promotions
generally have a 2-week duration and run from Tuesday to Monday.
• BIG W receives funding from vendors to support promotions. This funding is allocated
at a brand level. Predictions of vendor funding should be consistent with the supplied
historical data. See ‘scanback’ in the supplied data dictionary for related information.
• There must be at least two weeks without a promotion or price change before a new
promotion is activated.
Availability
• Not every article is sold in every store. You are provided with a daily count of how
many stores are selling each article (otherwise known as having the article “ranged”)
and a count of the stores both selling the article and with the article in stock.
• Stock availability for each article will vary by store. For the sake of simplicity, you don’t
need to consider constraints in supply that may result from price changes and
promotions. However, you may want to consider historical supply constraints whilst
developing your strategy.
Problem Description
You have been provided with a dataset from BIG W, relating to the sales and profit
performance for a selected set of Categories, along with data collected on competitor prices
for products directly competing with these BIG W products. The data provided spans
approximately two years and is detailed in the “Data Description” section below.
Please carefully review the “Mechanics of Pricing, Promotions and Availability” section above
to understand how pricing and promotions operate and what limitations exist on these
mechanisms.
In addition to this dataset, you are encouraged to explore external datasets to enrich
your analysis and feature engineering.
In this project, you will:
● As a business analyst, you will do a preliminary Exploratory Data Analysis (EDA) of
the dataset. You are expected to find or reveal all possible properties, characteristics,
patterns, and statistics hidden in the datasets. The results from your EDA may be
used for the final goal of identifying the metrics and relationships that are useful
for predicting daily sales volume (units sold) and, subsequently, gross profit.
● Synthesise your potential insights from the EDA and construct a model which can
predict the sales units and gross profit response to pricing and promotion
changes.
You will need to build a model with whatever machine learning approaches you feel appropriate.
You should evaluate your model/s on a range of metrics; however, the RMSE (defined below)
will be used to evaluate the performance of your final model on the test data. You should follow
an industry-recognised approach to Data Science problems (e.g. CRISP-DM) and include a
justification for your selected model. You will be required to show the methods you used to
prioritise your potential insights and defend the models and results with supporting evidence.
You will also be required to submit your sales predictions on the test data.
ൌ ඩ෍௡
௜ୀଵ
ሺෞ െሻ2
Important:
1) Please use the pre-split training and test set that has already been provided. Your
evaluation metrics on the test set are important.
2) Please consider which variables are not available at the time of fare look-up, and
exclude those as predictor variables (because in real life, your model won’t have them
available when making predictions!). You can read more about data leakage here:
https://www.kaggle.com/code/alexisbcook/data-leakage
● Based on your analysis, design a potential project for the Woolworths team to execute,
to take advantage of the sales, pricing, promotion and competitor data available to
increase the volume of units sold to their customers without decreasing the overall
gross profit.
This project could include (but is not limited to) a pricing strategy or optimisation of promotional
activity. Each project must be supported by an estimated improvement in overall volume
(units sold) and the corresponding predicted change in total gross profit with supporting
data and assumptions. Woolworths Group plans to run the project as a test over thirteen weeks,
from the week commencing Monday 05/02/2024 through to the week commencing Monday
29/04/2024 inclusive, so the group should focus on recommendations for this time period.
Please limit the number of recommended projects to 1-2. Also, note that it is ideal for
groups to recommend the deployment of their model. However, groups can also leverage
model insights for recommendations, as long as the recommendations are closely linked to
the insights and not overly general in nature (e.g. general app redevelopment or event).
Data Description
The data provided spans approximately two years and is detailed as below:
● BIG W Sales and Prices: contains ~1.85 million rows altogether. Provides daily sales
and profit data for each product sold in a selection of BIG W categories from February
2022 through to January 2024. The data has been pre-split into a training
(FinSalesPriceData_train.csv) and testing set (FinSalesPriceData_test.csv).
● Competitor Price Data: contains ~77.7k rows. Provides weekly level data on
competitor prices mapped to BIG W products from the week commencing Monday
14/11/2022 through the week commencing Monday 29/01/2024 (inclusive).
The provided data in both files has been limited to four product categories:
● Household Cleaning
● Baby Consumables
● Personal Hygiene
● Skin & Sun Care