Scribd
Upload
15 views

Marketing & Retail Analytics - Project 2

Marketing & Retail Analytics - Project 2

Uploaded by

amitkinwar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
15 views

Marketing & Retail Analytics - Project 2

Marketing & Retail Analytics - Project 2

Uploaded by

amitkinwar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 28

Marketing •Grocery Store's Data

& Retail
Analytics - •By: Amit Kumar
Project 2 •PGPDSBA
Contents:
• Executive Summary of the data
• Problem statement
• Executive Summary & Data Dictionary
• Assumptions about data
• Exploratory Analysis & Insights
• Weekly, Monthly, Quarterly, Yearly Weekday Trends in Sales
count
• Products counts & Year Wise top products
• Summary and Recommendations
• Market Basket Analysis
• Market Basket Analysis Meaning
• MRA KNIME WorkFlow & Qutput Table
• Associations Identified
• Association Rule Parameters
• MRA – values
• Association Rules Table
• Recommendation
• Recommendation
• Summary
Agenda
• Executive Summary of
the data
• Exploratory Data Analysis
• Market Basket Analysis
• Associations Identified
• Recommendations
01. Executive
Summary

• Problem statement
• Executive Summary
& Data Dictionary
• Assumptions about
data
Problem Statement
• A grocery store shared the transactional data with you. Your job is
to conduct a thorough analysis of Point of Sale (POS) data,
identify the most commonly occurring sets of items in the
customer orders, and provide recommendations through which a
grocery store can increase its revenue by popular combo offers &
discounts for customers.

• Data : dataset_group.csv
Executive Summary
• Objective: The project focuses on analyzing Point
of Sale (POS) data to identify popular product
combinations and discount opportunities that can
help a grocery store boost revenue by offering
attractive combo deals and discounts to customers.
• Dataset: 20641 Rows, 3 columns,
• Missing values : None
• Duplicate values: 4730
• The exploratory analysis and insights provide a
clear understanding of the data and highlight the
key trends and patterns in sales.
• Market Basket Analysis using association rules
was performed to identify the relationships between
the products purchased by the customers.
• This analysis helped to identify the products that
are frequently purchased together, which can be
used to create lucrative offers for the customers.
Assumptions:
• The data represents a list of items purchased at a grocery store on various
dates.
• Each entry in the data represents a single item purchased.
• The first column in the data represents the date the item was purchased.
• The second column represents the customer who made the purchase.
• The third column represents the item purchased.
• The same item can be purchased by multiple customers on different dates.
• There is no information provided about the quantity or price of each item.
• We have not dropped the duplicated values.
02.
Exploratory
Data Analysis

• Sold Products counts


• Trends in Sale count
• Summary and
Recommendations
Yearly Count
of Products
Sold
• We can see that the
data count is low for
2020 as we have
data till 26th Feb only.
Quarterly
Count of
Products Sold
• Highest sale was in
Q1 2019.
• Only Q1 data is
available in 2020.
• Sales in Q3 2018
best amongst all
Q3s.
Monthly
Count of
Products Sold
• In the Year 2019,
most of the products
were sold in March,
and the least were
sold in January.
• In 2018, most of the
products were sold in
January, and the
least were sold in
February.
Weekly Count
of Products
Sold
• Graph demonstrate
weekly product sales
count.
Weekday
Order Trend of
Products Sold

• Orders are more


frequent on
weekends and
Fridays, indicating a
possible rise in
demand toward the
end of the week.
Monthly Order
Trend of Products
Sold

• Order volumes vary by month and display


some seasonal fluctuations, with certain
months like January and May generally
higher.
Quarterly
Order Trend of
Products Sold

• There is a fairly
steady number of
orders across
quarters, with slight
increases in Q3 (July-
September).
Monthly
Count of
Products Sold

• We can see the


number of orders per
month in all three
years.
Number of
Products
Sold

• The pictures shows


the number of
product ordered.
Poultry leading the
basket.
Top 5
Products
Sold

• Here are the details


of top 5 products
sold.
Market • About Market Basket Analysis
Basket • Association rules and its relevance in this
case
Analysis • KNIME workflow image
(Associati • Threshold values of Support and
Confidence
on Rules)
About Market Basket Analysis
• Definition: Market Basket Analysis is a statistical method used to examine customer purchasing patterns and identify
relationships between products. This analysis helps businesses discover which products are commonly bought together
and gain insights into how customers' buying behaviors impact overall sales.

• Data: To perform market basket analysis, businesses require transactional data containing information like customer ID,
product ID, and transaction date. This data is used to build a matrix that maps the relationships between various
products, revealing patterns in customer purchasing behavior.

• Association Rules: Association rules help gauge the strength of relationships between products, using metrics like
support, confidence, and lift. Support indicates how often items appear together, confidence measures the likelihood of
buying one item given another, and lift shows the correlation strength between two items.

• Applications: Market Basket Analysis is widely used across industries like retail, e-commerce, and marketing. Retailers
leverage it to improve product placement and promotions, e-commerce platforms use it for personalized
recommendations, and marketers apply it to create targeted ad campaigns.
• Benefits: Market Basket Analysis boosts revenue by uncovering cross-selling opportunities and enabling targeted
promotions. It also enhances customer satisfaction through personalized recommendations, creating a better shopping
experience.
KNIME Workflow Image
• Association rules in Market Basket
Analysis uncover relationships and co-
occurrence patterns among items,
offering valuable insights into
customer buying behaviors and
Associatio preferences.
n Rules &
its • Association rules are valuable for
guiding businesses in optimizing
relevance. product placement, designing
targeted marketing campaigns, and
executing effective cross-selling and
upselling strategies, all of which
enhance customer satisfaction and
drive revenue growth.
Threshold values of
Support and Confidence

• The threshold value for Minimum Support is


0.05

• Threshold value for Minimum Confidence is 0.6

• In this analysis, we’ve set the Support and


Confidence thresholds to 0.05 and 0.6,
respectively. These values allow us to identify
significant and reliable association rules. By
applying these thresholds, we can filter out
less relevant rules and concentrate on those
with strong support and confidence, ensuring
that our analysis yields actionable insights for
informed decision-making.
Associati
on in
tabular
form
• Support: The support value indicates the
frequency or popularity of an itemset within the
dataset, showing how often a particular
combination of items appears in customer
transactions.

Support, • Confidence: Confidence measures the

confidenc probability that a customer who buys one item


will also buy another. It’s calculated as the ratio
of transactions where both items are purchased
e, & lift together to the transactions where only the first
item is purchased.

values • Lift: Lift indicates the strength of association


between two items, comparing the likelihood of
them being bought together to the likelihood of
independent purchases. A lift value above 1
suggests a positive association, meaning the
items are more likely to be bought together.
Suggestion of
Possible Combos
with Lucrative Offers
• Juice Combos: Given that juice has
high confidence and lift scores with
combinations including yogurt and
aluminum foil, offering discounts on
juice with these items could attract
purchases.
• Coffee/Tea Bundles: Since coffee/tea
combinations show good lift and
confidence with cereals and cheeses,
bundled offers on breakfast items
might enhance sales.
• Household Essentials Kit: Items like
laundry detergent and dishwashing
liquid appear frequently in
combination, suggesting a discounted
household essentials pack.
Recommendations
• Juice Combos: Given that juice has high confidence and lift scores
with Combo Deal: Introduce a combo offer where customers can
save on a bundle of yogurt, cereals, cheese, and juice, all priced at a
discounted rate when bought together.
• Buy Two, Get One Free: Run a "Buy 2, Get 1 Free" offer on juice,
yogurt, and popular snacks like chips or crackers, encouraging
customers to stock up on favorites.
• Bundle Promotion: Offer a “Household Essentials Bundle” with
dishwashing liquid, laundry detergent, and toilet paper at 20% off for
customers buying all three items.
• Cross-Selling Discount: Provide a discount on cereals, cheeses, and
bagels for customers purchasing coffee or tea, creating a well-
rounded breakfast deal.
• Limited-Time Discount: Run a special promotion offering 15% off
for customers purchasing poultry, mixes, and aluminum foil together,
ideal for easy meal preparation.
• These refined promotions encourage higher spending by tapping into
commonly bought item combinations.
Thank you! BY: Amit Kumar
PGPDSBA

You might also like