Retail Analytics

Module 4

Dr.Ilavarasi A K,VIT Chennai 1

Retail Analytics?
 Retailers use data analytics to improve inventory management,
marketing efforts, pricing, and product allocations.
 Retail analytics involves using software to collect and analyze data
from physical, online, and catalog outlets to provide retailers with
insights into customer behavior and shopping trends.
 It can also be used to inform and improve decisions about pricing,
inventory, marketing, merchandising, and store operations by
applying predictive algorithms against data from both internal
sources (such as customer purchase histories) and external
repositories (such as weather forecasts).
 In addition, retail analytics can measure customer loyalty, identify
purchasing patterns, predict demand, and optimize store layouts
 for instance, retailers can place items on store shelves that are
often bought together or offer personalized discounts to frequent
shoppers that will result in higher average basket sizes and more
frequent visits.

Dr.Ilavarasi A K,VIT Chennai 2

Retail Analytics Example
Planning production size

Dr.Ilavarasi A K,VIT Chennai 3

Retail Analytics Explained

 Retail analytics is the science of collecting,

analyzing, and reporting on data related to a
retailer’s operations.
 Retail analytics can apply to analyzing customer
behavior, tracking inventory levels, measuring the
effectiveness of marketing campaigns, and more.
 Analytics also helps retailers make better
decisions about which promotions to run and
which marketing strategies to focus on, as well as
when to staff up and down.
 Ultimately, data analytics helps retailers increase
sales, reduce costs, and improve customer
satisfaction and loyalty.

Dr.Ilavarasi A K,VIT Chennai 4

 Benefits of Retail Analytics
 Reducing stockouts and the need for discounts: Retail analytics helps users
understand demand trends so they can have enough product on hand
◦ For example, analytics can help determine how quickly demand falls for fashion
items that are driven by the popularity of social influencers.
 Improving personalization: Analytics helps retailers understand their
customers’ preferences and thus capture more demand than their competitors.
◦ For example, using purchasing history, a book retailer can alert customers who
have shown interest in American history when a new book by historian Ron
Chernow becomes available for preorder.
 Improving pricing decisions: Data analytics can help retailers set the optimal
prices for their goods. Retailers can thus maximize profits by avoiding setting
prices higher than the market will bear or lower than what customers would be
willing to pay.
 Improving product allocations: Analytics can help retailers decide how to
allocate products in different geographic regions, distribution centers, and stores,
reducing needless transportation costs.
Dr.Ilavarasi A K,VIT Chennai 5
Types of Retail Data Analytics

 Descriptive analytics reflect and

explain past performance;
 Diagnostic analytics to determine the
root cause of a given problem;
 Predictive analytics to forecast future
results;
 Prescriptive analytics to recommend
next steps.

Dr.Ilavarasi A K,VIT Chennai 6

Descriptive analytics
 Descriptive analytics is the foundation for
more sophisticated types of analytics, It
addresses fundamental questions of “how
many, when, where, and what”—the stuff
of basic business intelligence tools and
dashboards that provide weekly reports
on sales and inventory levels.

Dr.Ilavarasi A K,VIT Chennai 7

Diagnostic analytics
 Diagnostic analytics helps retail
organizations identify and analyze issues
that may be hindering their performance.
 By combining data from multiple sources,
such as customer feedback, financial
performance, and operational metrics,
retailers gain a more comprehensive
understanding of the root causes of
problems they face.

Dr.Ilavarasi A K,VIT Chennai 8

Predictive analytics

 Predictive analytics helps retailers anticipate

future events based on several variables,
including weather, economic trends, supply
chain disruptions, and new competitive
pressures.
 This approach often takes the form of a
what-if analysis, which, for example, would
let a retailer map out what would happen if
it offered a 10% discount versus 15% on a
product, or estimate when it would run out
of stock based on a given set of possible
actions.

Dr.Ilavarasi A K,VIT Chennai 9

Prescriptive analytics

 Prescriptive analytics is where AI and big

data combine to take those predictive
analytics outcomes and recommend actions.
 Prescriptive analytics can, for example,
provide customer service agents with
suggested offers they can pass to customers
on the fly, whether that be an upsell based
on previous purchase history or a cross-sell
to satisfy a new customer inquiry.

Dr.Ilavarasi A K,VIT Chennai 10

 How Is Retail Analytics Used?
 Companies use retail analytics to explain past operational and
financial performance, diagnose what might have gone wrong,
suggest alternative approaches that would have been more
productive, forecast demand, and offer suggestions
 In-store analytics tools use data generated from POS systems
and in-store video cameras to help retailers analyze customer
shopping patterns so they can place products more effectively
 Customer analytics uses data from systems that customers
interact with, including POS systems, websites, phone logs, and
customer service chats. Analyzing this data helps retailers
determine Popularity of items and cause of returned items
 Inventory analytics, as the name suggests, assesses inventory
levels for goods a retailer has on offer. Inventory analytics can,
for example, reduce the labor and shipping costs associated with
carrying too much safety stock.

Dr.Ilavarasi A K,VIT Chennai 11

 Contd..
 Merchandise analytics helps retailers determine whether they’re displaying
their wares effectively, mostly in physical stores, to increase profit margins
across products
 Web analytics tracks the digital footprint of consumers as they linger over
certain parts of a web page or click from one page to another
 Business intelligence (BI) reports, often presented in the form of
dashboards, are preset to show certain key performance indicators, such
as inventory turns and sell-through rate. They are used mainly to share
top-line trends with peers and senior management.
 Demand forecasting forecasts demand for particular items sold online
based on the path customers followed to view those items, move them to
their shopping cart, remove those items, or abandon the cart entirely.
 Sales forecasting helps retailers predict future sales based on actual sales
figures and other factors. Used in tandem with demand forecasting, it can
predict what total demand will be for an item across all channels

Dr.Ilavarasi A K,VIT Chennai 12

Retail Analytics Tools

 Point-of-sale systems: These are the systems

that retailers use to track and manage customer
transactions. POS systems provide data on
customer purchases and can generate reports on
sales and customer trends.
 Customer relationship management
(CRM) software: This software category
includes applications that manage sales, marketing,
customer service, and ecommerce processes.
Retailers use these applications to track
interactions with customers, retain data about
individual customers.

Dr.Ilavarasi A K,VIT Chennai 13

 Business intelligence tools: Retailers use BI tools to
synthesize information gleaned from large volumes and
different sets of data, mostly to track key performance
indicators such as customer loyalty, inventory turns, sell-
through rate, and days on hand.
 Inventory management systems: Retailers use this
software to track items in stock, monitor inventory
levels in warehouses and distribution centers, and create
forecasts of demand. It also helps retailers identify
optimal locations for storing certain items to minimize
transportation expenses and ensure that goods are
available to meet customer demand.
 Predictive analytics: This type of analytics uses data
from prior transactions, communications, and other
actions to predict future trends and behaviors.
Dr.Ilavarasi A K,VIT Chennai 14
CHURN
 Churn prediction involves identifying at-risk customers who are
likely to cancel their subscriptions or close/abandon their accounts.
 A churn model works by passing previous customer data through a
machine learning model to identify the connections between
features and targets and make predictions about new customers.
 Identifying churn before it happens can help businesses take
proactive action to retain customers. This includes targeted re-
engagement campaigns, personalized customer education, and
more.
 The first step to creating a churn model is to collect relevant data,
including product usage data and direct feedback data from
customer surveys.
 Next, step is to analyze trends in the data to find the main reasons
behind customer churn.
 Finally, pass the data through a logistic regression algorithm (such as
the random forest algorithm) to identify key data points and make
future predictions.

Dr.Ilavarasi A K,VIT Chennai 15

Understanding the churn
prediction model
 A churn prediction model is a machine learning model that predicts
whether a customer will likely churn. At a high level, predicting
customer churn requires a detailed understanding of your
customers.
 This understanding is derived by examining the historical data of
your customers. A good churn prediction dataset will include
multiple predictive features that describe your customer – contract
type, subscription price, etc.
 It should also have a target variable (the feature you want to
predict). In this case, this will be a column indicating whether the
customer churned or not.
 Finally, you’ll need a machine-learning model (specifically a logistic
regression algorithm like decision trees, random forest, SVM, or XG
Boost) to find patterns in the data and make accurate predictions.
 To summarize:
Historical data + machine learning = churn model

Dr.Ilavarasi A K,VIT Chennai 16

Why is customer churn
prediction important?
 Churn is expensive. The cost of any new customer
acquisition is always higher than the cost of retaining existing
customers. This is especially true for SaaS companies with
the subscription business model.
 Therefore, predicting customer churn before it happens is an
important part of modern business management. It helps
marketing teams to:
 Provide more targeted re-engagement campaigns for at-risk
customers.
 Create more focused customer education content to
increase customer lifetime value.
 Retain customers before they churn.
 On a larger scale, churn trends can help marketers build
customer personas to target a market segment with better
messaging and boost customer acquisition.

Dr.Ilavarasi A K,VIT Chennai 17

Creating churn prediction
models to prevent churn
 There are three main steps to creating a
customer churn prediction model. They are:
 Data preparation: This involves gathering
relevant data and preparing it for use in your
model. It is sometimes said that data preparation
forms 80% of data scientists’ jobs.
 Exploratory data analysis: This step aims to
understand your data and discern the factors
behind customer churn in your business.
 Predictions: The final step of your data science
process involves creating a predictive model to
identify high-risk customers before they churn.
Dr.Ilavarasi A K,VIT Chennai 18
Creating a churn model for
business
 Leverage data points for predicting
customer churn
The first step to creating your model is collecting
the right data. The more data you have, the more
accurate your predictions will be. Consider some
data collection methods for a churn model.
 Monitor product usage data of existing
customers
Product usage data tells you how and when your
customers are using your product. It reflects how
customers use your software, capturing their
engagement and behavioral data.
Dr.Ilavarasi A K,VIT Chennai 19
 Some important product usage data points for your
model include:
 Feature usage data: How are users engaging with the
different features of your product? This metric reveals the
most popular/relevant feature of your product.
 Customer behavior: Customer behavior data captures
everything a user does within your product. This includes
when they use your product, how long they use it, which
features they engage with, how they progress through the
product, etc.
 Clicks: This is a record of the number of times a user clicks
or interacts with a UI element, such as a button, checkbox,
text area, menus, etc.
 Others: Other product usage data you can track include
time-to-value, product stickiness, interactions, etc.

Dr.Ilavarasi A K,VIT Chennai 20

User pilot in churn prediction
and prevention
 Userpilot is a product growth and customer analytics platform that
helps you collect, analyze, and act on customer data in your
product. Thanks to its no-code nature, you don’t need a data
science team to work with it.
 Some key features include:
 Feature tagging: Tag product features and UI elements to track
how they’re used and why users interact with them.
 Surveys: Create and launch in-app surveys to collect customer
data directly. This includes everything from customer success
surveys like NPS and CSAT to general feedback surveys.
 Segmentation: Segment users based on their session data,
feature usage data, feedback, and more, and tailor their experiences
to help them get the most out of your product.
 Funnel analysis: Funnels in Userpilot will allow you to break
down the data for detailed analysis of the conversion rates.

Dr.Ilavarasi A K,VIT Chennai 21

Funnel Chart

Dr.Ilavarasi A K,VIT Chennai 22

Real estate dataset

Dr.Ilavarasi A K,VIT Chennai 23

Google spreadsheet implementation

Dr.Ilavarasi A K,VIT Chennai 24