INTRODUCTION FOUR ERAS IN TEN YEARS A REVOLUTION IN ANALYTICS Flashcards
What serious threat does complacency pose to businesses?
Complacency can lead to misguided satisfaction and failure to recognize competitive threats.
Examples include photographic film companies, newspapers, and movie rental companies that failed to adapt to new technologies.
Who is Jim Casey and what concept did he advocate?
Jim Casey was the founder of UPS and advocated for ‘constructive dissatisfaction.’
He believed in continuously restructuring and reinventing UPS to counter competitive threats.
What significant advantage did UPS gain from adopting analytics?
UPS gained the ability to continuously assess and improve every facet of their business.
This included the design of handheld devices for drivers and building a large data warehouse.
What is ORION and its significance for UPS?
ORION is a prescriptive analytics model that optimizes delivery routes for drivers, saving time and fuel.
It generates over $400 million in annual cost savings for UPS.
What was the primary data storage solution during the Analytics 1.0 era?
The primary data storage solution was the relational data warehouse.
It required data to be structured in rows and columns and involved a lengthy ETL process.
Fill in the blank: Analytics 1.0 was heavy on _______ analytics.
descriptive
What cultural challenges did organizations face during the Analytics 1.0 era?
The culture was reactive and slow, with analytics often being used only to support decisions rather than drive them.
Many decisions were still made based on intuition rather than data.
What is the key difference between data analysts and data scientists?
Data scientists are more involved in guiding strategic decisions and product development rather than just supporting decision-making.
They often prefer to work closely with executives and focus on creating data-driven products.
What does the term ‘self-service analytics’ refer to?
Self-service analytics refers to tools that allow users to create their own reports and visualizations without needing analytical professionals.
This trend emerged to make analytics more accessible.
What major change occurred in the analytics landscape after 2007?
The emergence of Analytics 2.0, 3.0, and 4.0, which introduced new ways of handling and analyzing data.
This shift was characterized by the rise of big data and advanced analytical technologies.
What was a significant technological development in the era of Analytics 2.0?
The creation of Hadoop, an open-source program for storing large amounts of data across distributed servers.
Hadoop allows for minimal processing and is a cost-effective way to manage big data.
True or False: The analytics created during Analytics 2.0 were typically sophisticated.
False
They were often less sophisticated, focusing on flexibility and low cost.
What is the relationship between data scientists and senior executives in modern analytics?
Data scientists often work closely with senior executives, guiding strategic decisions rather than remaining in the back office.
They seek to have a direct impact on business outcomes.
What is the ‘big data equals small math’ syndrome?
It refers to the phenomenon where big data is analyzed using relatively simple mathematical techniques.
This was noted by data scientists who observed that the analysis was not as complex as one might expect.
Fill in the blank: The analytics era prior to 2007 is referred to as _______.
Analytics 1.0
What was the focus of companies competing on analytics in the mid-2000s?
They focused on improving decision making and performance using available analytical capabilities.
Despite challenges, they were dedicated to making analytics work.
What role did the relational data warehouse play in Analytics 1.0?
It served as the primary data storage solution, requiring structured data for analysis.
This approach had its challenges, including difficulties in data management.
What was the career path of the individual mentioned in the text?
Data scientist at LinkedIn, venture capital, head of product at a startup, to the White House
The individual admitted to having an office in the basement of the White House.
How did data scientists interviewed perceive decision support?
Many found it uninteresting, with one referring to it as ‘the Dead Zone’
They preferred to focus on products and features.
What are some data products developed by LinkedIn?
- People You May Know
- Jobs You May Be Interested In
- Groups You Might Like
These products contributed to LinkedIn’s growth.
Which company acquired LinkedIn and for how much?
Microsoft for $26 billion
What is a key lesson from Analytics 2.0 practitioners?
Analytics are core to the strategies of many firms, competing on analytics more than others.
What is the motto of Facebook regarding its experimental culture?
‘Move fast and break things’
What does Analytics 3.0 represent?
A combination of big data and small data for big companies
What types of data do companies analyze in the Analytics 3.0 era?
- Clickstream data
- Social media sentiments
- Sensor data from IoT
- Customer location information
- Past purchase data
This era emphasizes the integration of all data types.
What is ‘operational analytics’?
Analytics integrated into production processes and systems
This allows real-time offers and supply chain optimizations.
What is the ORION project and its significance?
A UPS project for driver routing that saves about half a billion dollars annually
It demonstrates the scale of analytics in the 3.0 era.
What new business model has GE developed?
A ‘digital industrial’ business model powered by sensor data
This model predicts service needs based on data.
What is the estimated annual revenue generated by United Healthcare’s Optum?
$67 billion
What characterizes Analytics 4.0?
The rise of autonomous analytics, reducing the role of human analysts
What technological force is viewed as disruptive in Analytics 4.0?
Artificial intelligence or cognitive technologies
What statistical model is commonly used in cognitive technologies?
Machine learning
What is a key advantage of machine learning models?
They can create thousands of models quickly compared to traditional methods
What is the ‘black box’ problem in machine learning?
The difficulty of interpreting models generated by machine learning
What is propensity modeling?
Modeling customer likelihood to buy products
This can be done with traditional or autonomous analytics approaches.
How many propensity models did the company increase to after using machine learning?
From 150 to about 5,000 models
What percentage of models produced by Modern Analytics are generated without human intervention?
95%
What does the transition to machine learning require from analytics experts?
New skills in assembling data and monitoring outputs
What is a challenge companies face when interpreting machine learning models?
It is difficult to fully interpret models with thousands of variables
What type of computing power is essential for deep learning?
High-level computing power and large amounts of data
What is the role of cloud computing in Analytics 4.0?
Provides virtually unlimited computing capability at reasonable prices
What is the main goal of leading analytical organizations in the Analytics 4.0 era?
To leverage cognitive technologies and machine learning for data handling
What is the transition from traditional analytical methods to Analytics 4.0?
A move from artisanal analytical methods to autonomously generated models
What are organizations required to integrate to compete on analytics?
New technologies and new methods
What is a key characteristic of organizations wanting to compete on analytics?
They need to be very nimble
What type of technologies is Capital One using for cybersecurity, risk, and marketing?
Cognitive technologies
What skills are required for doing Analytics 1.0?
- Statistics
- Data integration and cleaning
- Understanding of the business
- Effective communication about data
- Inspiring trust among decision-makers
What new skills are required in the Analytics 2.0 era?
- Experimentation capabilities
- Transforming unstructured data
- Familiarity with open-source development tools
- Knowledge of product development and engineering
- Familiarity with visual analytics
What is a significant challenge in implementing operational analytics?
Change management skills are needed
What are some technical skills involved in Analytics 4.0?
- Machine learning
- Deep learning
- Natural language processing
- Work design skills
What percentage of executives felt their big data initiatives were successful according to the 2016 survey?
80.7 percent
What impediment did 43% of executives mention regarding big data initiatives?
Lack of organizational alignment
What percentage of firms reported that they had created a data-driven culture successfully?
37 percent
What is necessary for culture change in organizations?
Committed leadership
What do established firms struggle with compared to digital startups?
Mastering digitally driven, analytically rich strategies
What does the DELTA model address?
Factors an organization must address to improve at analytics
What is a significant aspect of the book’s content?
Descriptions of how analytics can lead to better business performance
What does Part II of the book focus on?
How-to guide for competing on analytical capabilities
What new content has been added to the revised edition of the book?
- Data scientists and their roles
- Big data’s impact on analytics
- Open-source software like Hadoop
- Data products based on analytics
- Machine learning technologies
- Internet of Things implications
- Cloud computing architectures
- Embedding analytics in operational systems
- Visual analytics
True or False: The challenges of developing an analytical culture have significantly changed since 2007.
False
