E-commerce refers to the buying and selling of goods and services online. Analytics in e-commerce helps businesses optimize their websites, grow their customer base through marketing campaigns, and retain customers through predictive analytics. Predictive analytics can help e-commerce businesses enhance product recommendations, pricing, and supply chain management. A case study shows how Amazon plans to use predictive analytics for anticipatory shipping to ship items before orders are placed. Another case study outlines how Mu-Sigma helped a CPG company improve revenue management through a strategic revenue management platform integrating data from multiple sources. The future of e-commerce is expected to see continued growth in online sales and mobile commerce.

1. Analytics in E-commerce
Shallote Dsouza

2. About E-commerce
 E-commerce is the activity of electronically buying or selling of products on online
services or over the Internet.
 Electronic commerce draws on technologies such as mobile commerce, electronic funds
transfer, online transaction processing, electronic data interchange (EDI), inventory
management systems, and automated data collection systems
 E-business refers to all aspects of operating an online business, E-commerce refers
specifically to the transaction of goods and services.

3. E-commerce in India

4. Verticals of E-commerce
 BUSINESS-TO-BUSINESS (B2B)
B2B e-commerce refers to all electronic transactions of goods and sales that are conducted between
two companies
 BUSINESS-TO-CONSUMER (B2C)
B2C e-commerce deals with electronic business relationships between businesses and consumers.
Many people enjoy this avenue of e-commerce because it allows them to shop around for the best
prices, read customer reviews and often find different products that they wouldn’t otherwise be
exposed to in the retail world
 CONSUMER-TO-CONSUMER (C2C)
This level of e-commerce encompasses all electronic transactions that take place between consumers.
Generally, these transactions are provided by online platforms (such as PayPal), but often are
conducted through the use of social media networks (Facebook marketplace) and websites (Craigslist)

5.  CONSUMER-TO-BUSINESS (C2B)
C2B e-commerce is when a consumer makes their services or products available for companies to
purchase. An example of this would be a graphic designer customizing a company logo or a
photographer taking photos for an e-commerce website
 BUSINESS-TO-ADMINISTRATION (B2A)
This e-commerce category refers to all transactions between companies and public administration. This
is an area that involves many services, particularly in areas such as social security, employment and
legal documents
 CONSUMER-TO-ADMINISTRATION (C2A)
Another popular e-commerce category, C2A e-commerce encompasses all electronic transactions
between individuals and public administration. Examples of this include taxes (filing tax returns) and
health (scheduling an appointment using an online service)

6.  Ecommerce analytics simply refers to any tool or strategy designed to analyze large
amounts of data in order to produce actionable insights
 Because it exists in an almost entirely virtual space, ecommerce generates complex,
comprehensive datasets — particularly those related to client behavior
 More data was created in 2017 than was created during the previous 5,000 years combined.
That is a lot of data to measure, parse, and analyze
 Hence to add value to business analytics is important
Why Analytics in E-commerce

7.  Growing the customer base with SEO, SEM, and email campaign insights
• Nearly as nearly 40% of online shoppers begin their shopping experience by performing an
online. As a result, any business that is not focusing on search engine optimization (SEO) is
missing out on a huge chunk of the consumer base
• Paid search engine marketing (SEM) campaigns can edge players ahead of the competition
and help them win more shoppers
 Reaching new customers is important, but holding onto them is crucial
• New customer acquisition may give business legs, but customer retention is truly the
backbone of a brand’s sustained success
• Acc. to experts 80% of future revenue will come from 20% of the existing clients
How Analytics helps in E-commerce

8.  Optimizing website with data-driven insights
• Ecommerce analytics can provide us with data-driven insights into how shoppers
interact with a site — both the good and the bad
• These insights take the guesswork and subjectivity out of website optimization, and
uncover opportunities for improvement and innovation
 Capture customer data no matter how or where they shop
• Ecommerce analytics must also evolve to track shoppers wherever they may be,
whether they are purchasing a product on Instagram, discovering a brand on their
phone, or cashing in a gift card at a pop-up shop.

9. Predictive Analytics in E-commerce
 Predictive Analytics provides e-commerce businesses with a deeper understanding of customer
habits and preferences by analyzing their past click-through behaviors, shopping history, and
product preferences, in real time
 E-commerce businesses can harness the potential of predictive analytics to offer enhanced
product recommendations and promotions
 Predictive analytics enables e-commerce businesses to enhance pricing models
 Predictive analytics helps e-commerce businesses to minimize fraud by learning which product
categories are most susceptible to fraud and manage them accordingly
 It offers effective supply chain optimization opportunities for e-commerce businesses
 Predictive analytics allows e-commerce businesses to make critical business decisions faster

10. Case Study – Amazon Wants to Use Predictive Analytics to
Offer Anticipatory Shipping
 Amazon will ship customer’s orders before they’ve been placed with their newly patented
“Anticipatory Shipping”
 How Will Anticipatory Shipping Work?
 Paired with predictive analytics tools and a massive trove of Amazon customer data, the
anticipatory shipping process will ensure popular items remain in an effective limbo to cut
down on fulfillment times
 When customers in a particular area order a product, it will be sent from a shipping hub — or
where it’s stored on nearby trucks — in a much shorter timeframe. The goods may even be
stored on pallets in smaller, strategically placed warehouses until customers are ready to
order
 The company will base locally stored inventory on area-specific stats such as previous orders,
basket or shopping cart contents, customer wish-lists and even popular regional listings. It’s
likely it will also integrate with Amazon’s Alexa platform, which allows you to order products
automatically or on a schedule.
 Think of receiving toilet paper or paper towels shortly after you inform Alexa that you’ve run
out. The only difference is the products reach you much faster than if they were leaving a
conventional fulfillment center

11. Case Study- Strategic Revenue Management by Mu-
Sigma
 The Problem
One of the largest global consumer packaged goods (CPG) manufacturers in the food sector was
struggling to decode the real drivers of sales and leverage them for better returns on revenue
management
 They were unable to root basic trade promotion questions in reliable data:
• How much should be spend on trade?
• Which levers to use?
• Which promotions to invest in?
 The Mu Sigma Approach
Mu Sigma uncovered some systemic problem triggers:
 • Fragmented data: Their data was being lodged into many systems in different formats, thus
under-contributing to trade spend insights simply for the lack of unified data architecture
 • Unilateral outcomes: Managers with different retail partners followed diverse business
outcomes and KPIs

12.  Mu Sigma worked on the following
• Promotions: When to promote a product?
• Strategic Pricing: At what price point/discount should it be promoted?
• Trade Architecture: Should the promotion be supported by features like ads and on which
• Portfolio mix: If two products in the portfolio are co-promoted, will one cannibalize the sales of
other?
 The Solution
Since multiple systems were the major culprit behind this product brand’s ineffective revenue
management strategy, they formed a multi-disciplinary team to institute a flexible Strategic Revenue
Management (SRM) framework
The Strategic Revenue Management Platform was created with following as foundation:
Data Foundation
Data from multiple data sources were integrated into a promotion-specific cloud-based data lake

13. Analytical Foundation
The analytical layer deployed machine learning-based models for:
• Sales Driver Analysis: This involved building models to understand the exact contribution of each driver to the
final sales of a product
• Optimization and Scenario Builder: The sales driver models coupled with an optimization engine that
all pillars of trade promo and suggests the most optimal promotional calendar for each retailer and product
combination
Insight Consumption
The consumption layer hosts a suite of flexible planning and analysis tools with visualization, simulation, and
optimization capabilities which enable:
• Insights Generation
• Insights Automation
• Third-Party Tools
• Third-Party Analysis
The Impact
The cumulative insights helped identify opportunities for trade promotion optimization of about $15M yearly
in the USA alone

14. Future growth of E-commerce
 By 2040, around 95% of all purchases are expected to be via ecommerce
 The number of online buyers is expected to reach 2.14 billion by 2021
 The total value of global retail ecommerce sales will be nearly $4.88 trillion in 2021
 Online stores with a loud social media presence will get 32% more sales on average
than those who do not
 On average, 52% of ecommerce businesses already have Omni channel capabilities
 The shares of m-commerce in all ecommerce is predicted to reach to 72.9% by 2021
 The number of mobile users worldwide is forecasted to reach 7.26 billion in 2020. By
2021, mobile is predicted to dominate online sales. Probably, 73% of ecommerce
sales will be made on mobile in 2020
 The new trend of Ecommerce has become a game-changer for both B2B and B2C
companies. It is being conceived that B2B Ecommerce is outpacing B2C, and will soon
outperform it

15. Recommendations
 Improve user experience to increase engagement
 Better working networks as ecommerce is totally dependent on the internet
 Preparations to handle worst case scenarios example, Covid-19 pandemic
 Improve the existing recommendation systems
 Improve the existing delivery systems

16. Thank You