This document discusses big data, including its drivers, characteristics, use cases across different industries, and lessons learned. It provides examples of companies like Etsy, Macy's, Canadian Pacific, and Salesforce that are using big data to gain insights, increase revenues, reduce costs and improve customer experiences. Big data is being used across industries like financial services, healthcare, manufacturing, and media/entertainment for applications such as customer profiling, fraud detection, operations optimization, and dynamic pricing. While big data projects show strong financial benefits, the document cautions that not all projects are well-structured and Hadoop alone is not sufficient to meet all business analysis needs.
2. What is Big Data
More Devices
More
Consumption
More Content
New & Better
Information
Big Data encompasses not only
the content itself, but how
it’s consumed.
*Source: IDC 2011
Every gigabyte of stored content can generate a
petabyte or more of transient data*
The information about you is much greater than
the information you create
Big Data Drivers:
The high adoption of data capture
and creation technologies
Increased “interconnectivity” drives
consumption, creates more data
Inexpensive storage makes it
possible to keep more data for
longer period
Hadoop software and analysis tools
turn data into information
6. Companies using Big Data
Company : Etsy
Category : On line retailer
Big Data Attribute : Volume
Revenue: $895 M
Doing Large scale analysis of clickstream data,
company is discovering important product attributes
for the user like Materials, prices, textures etc. They
use these attributes to rank the products in search
results.
Percentage of users making purchases increased
7. Companies using Big Data
Company : Macy
Category : Brick and Mortar retailer
Big Data Attribute : Volume and Speed
Revenue: $26 B
Price check analysis of its 10,000 articles across 800 stores
nationwide in less than 2 hours. When ever a neighboring
competitor between New York and Los Angles goes for
aggressive price reductions, Macy’s follows the suite. If there
is no competition, price remains unchanged. There are
around 270 million different prices across entire range of
goods and locations. Just completing this analysis at this
speed was unthinkable without Big Data.
Reduce loss to local competition
8. Companies using Big Data
Company : Canadian Pacific (Using GE System)
Category : Brick and Mortar retailer
Big Data Attribute : Volume and Velocity
Revenue: $5.7 B
“Trip Optimizer” is a fuel-saving system that GE has
developed for freight trains. It takes into account data such as
track conditions, weather, the speed of train, GPS data and
“train physics”, and makes decisions about how and when the
train should break.
Reduced fuel usage by 4 - 14 %
9. Companies using Big Data
Company : Sears Holdings (Sears and Kmart)
Category : Brick and Mortar retailer
Big Data Attribute : Volume and Velocity
Revenue: $42 B
Sears’ process for analyzing marketing campaigns for loyalty
club members used to take six weeks on mainframe,
Teradata, and SAS servers. The new process running on
Hadoop can be completed weekly. For certain online and
mobile commerce scenarios, Sears can now perform daily
analyses. What's more, targeting is more granular, in some
cases down to the individual customer. Whereas the old
models made use of 10% of available data, the new models
run on 100%.
Part of a five-part strategy to get the company back on track
10. Big Data – Case Study
Company : Salesforce dot com
Category : Software Vendor
Big Data Attribute : Volume and Varieties
Revenue: $2.27 B
What are they solving
• Track feature usage/adoption across 130k+ customers
examples: Accounts, Contacts, Visualforce, Apex,…
• Track standard metrics across all features
examples: #Requests, #UniqueOrgs, #UniqueUsers, AvgResponseTime,…
• Track features and metrics across all channels
• Example –API, UI, Mobile
• Primary audience: Executives, Product Managers
18. Use Cases by Verticals
Financial Services
Customers Insights – using Hadoop to improve customer profile analysis to help
determine Eligibility for equity, capital, insurance, mortgage and credit
Fraud detection – Hadoop provides the scalable method to easily detect many
Types of fraud and loss prevention. Companies are also developing models to predict
Future fraud events like PayPal.
Micro Targeting – Banks have multiple silo systems for loans, mortgages,
investments. Hadoop can be used to provide aggregated view on customer
profitability.
Healthcare and Life Science
Gene Sequencing – The sequencing of DNA for organism holds huge promise for
human kind.
Health Information exchange – Hadoop can be used by providers to manage and
share healthcare records from mixed data sources such as images, treatments,
demographics and billing
19. Use Cases by Verticals
Manufacturing
Service Management – The availability of sensors and corresponding ability to
effectively store and analyze large data feeds across customer locations and product
SKUs, has resulted in more effective and efficient service.
Operations – Hadoop can also improve the post sales maintenance process. The
manufacturing industry is adding sensors to equipment to collect much more data
on the operations of the equipment. Collecting and analyzing these data improves
the maintenance process, increase productivity and reduces cost.
Media and Entertainment
Price analytics – Companies can use Hadoop to determine dynamic pricing for
everything from game tickets, web bases games to music and videos.
Customer Insights – Media companies need to better understand market segments
and consumer personal preferences and behavior to better match each brand to its
segment and help increase sales.
20. Big Data Lessons
Sources
1. Survey of 752 corporate executives by SAS institute from a broad range of
sectors and countries
2. Interviews of 17 data pioneers
3. Information Week Survey of CIOs
Pros
• Strong link between financial performance and effective use of big
data
• Social media analytics and web tracking technologies can transform
the way business collect data about their customers
Cons
• Too many big data projects are structured like boil-the-ocean
experiments
• Hadoop-based techniques aren't enough to meet business needs for
analysis