Lecture_3 Flashcards
When we start to map value in data ecosystem, what is the first thing need to be considered?
The value of data. How unique the data is and how to use the data by whom.
What are the potential problems as the data ecosystem has been evolving?
Raw data and actual application of data-derived insights are totally different.
What is non-rivalrous nature of data?
One of the uniqueness of data asset. It means data can be sued by multiple parties simultaneously.
What is sheer-diversity
one of the uniqueness of data asset. Including data type,
What is the data use case for cost & revenue optimization?
Predicitve maintenance, product improvements
What is the data use case for marketing & advertising
By analyzing customer transactional & behavior data from multiple sources.
What is the data use case for marketing intelligence?
Data is used to deliver strategic insight such as where does the next campaign occur.
What is the data use case for makret-making?
Use data to match clients’ needs and develop efficient matching.
What is the data use case for AI training
Machine learning requires hug quantities of training data.
Five steps for a functional ecosystem.
- data generation & collection.
- data aggregation.
- data analysis
- data infrastructure
What is data genration role?
Source and platform where data are initially captured.
What is data aggregation?
Process and platforms for combining or clean up the data from multiple sources.
What is data analysis?
Visualization data. Collect insight from the data.
What is data infrastructure?
Involving hardware & software
When do factors drive data value up at data genration stage?
Certain data type will have higher value if collection barries are extremely high or data can’t be legally shared between parties.
When do factors drive data value down at data generation stage?
Growth in available proxies and expansion of open access will increase suuply.
When do factors drive data value up at data aggregation stage?
More applications being developed and aggregation process is technically challenging or requires a neutral third party
When do factors drive data value down at data aggregation stage?
Technology makes data aggregation easier.
When do factors drive data value up at data analysis stage?
Talent shortage.
Deep sector expertise needed.
Close relationshiop to actual use or implementation clarifies value.
When do factors drive data value down at data analysis stage?
Scope could be limited as solutions will be for vertical applications.
The data generation and collection value will go down, while data analysis value will go up and data aggregation will remain the same.
True.
Credit card application ecosystem
See slide 17
What drive data collection?
Supply and demand
As supply of data wil expand, the generation of raw data will become less valuable.
True
In supply site, the market is shaped by
Difficulty of collection, access and availability of substitute data
On the demand side, the market is shaped by
Easy of use, network effects, and value of the ultimate uses of data.
When we need a disruptive model?
Assets are underutilized.
Supply/demand mismatch.
Dependence on large amounts of personalized data.
Large value in combining data from multiple sources.
R&D is core to the business model.
Decision making is subject to human biases.
Speed of decision making limited by human constraints.
Large value associated with improving accuracy of prediction.
Orthogonal data can be used in
Business model such as insurance and health care companies.
Example of orthogonal data usage
Usually health care companies relied on medical histories and examination, but with leveraging telematics data, news sets of orthogonal data can gain insights into customer behavior.
What is hyperscale real-time matching?
Hyperscale digital platforms can use data and analytics to meet both types of needs and have notable impact when:
- demand and supply fluctuate frequently.
- Poor signaling mechanisms produce slow matches.
- Supply side assets are under utilized.
What is the use case for hyperscale real-time matching?
Transprotation disruption. Use supply and demand fluctuate
Use case for radical personalization
Tailored offers for different customers based on their preference and characteristics
What is the first step in creating value from data?
Ensure access to all relevant data.
How to define data lake?
Integrate all data from structured, unstructured to internal and external
Analytics can improve four aspects of decision making.
- Speed/adaptability
- Accuracy.
- Consistency/reliability
- Transparency
Data and analytics can overcome human limitations in decision making
True
Archetype of disruption
- Business models enabled by orthogonal data.
- Hyperscale, real-time matching.
- Radical personalization.
- Massive data intergration capabilities.
- Data-driven discovery.
- Enhanced decision making.
What is the use case for data-driven discovery?
Data and algorithms can enhance innovation process.