3.14: Finding Value in Data- Business Intelligence

With the rise of Big Data and a myriad of new tools and techniques at their disposal, businesses are learning how to use information to their advantage. The term business intelligence is used to describe the process that organizations use to take data they are collecting and analyze it in the hopes of obtaining a competitive advantage. Besides using their own data, stored in data warehouses (see below), firms often purchase information from data brokers to get a big- picture understanding of their industries and the economy. The results of these analyses can drive organizational strategies and provide competitive advantage.

Data Visualization

Data visualization is the graphical representation of information and data. These graphical representations (such as charts, graphs, and maps) can quickly summarize data in a way that is more intuitive and can lead to new insights and understandings. Just as a picture of a landscape can convey much more than a paragraph of text attempting to describe it, graphical representation of data can quickly make meaning of large amounts of data. Many times, visualizing data is the first step towards a deeper analysis and understanding of the data collected by an organization. Examples of data visualization software include Tableau and Google Data Studio.

Data Warehouses

As organizations have begun to utilize databases as the centerpiece of their operations, the need to fully understand and leverage the data they are collecting has become more and more apparent. However, directly analyzing the data that is needed for day-to-day operations is not a good idea; we do not want to tax the operations of the company more than we need to. Further, organizations also want to analyze data in a historical sense: How does the data we have today compare with the same set of data this time last month, or last year? From these needs arose the concept of the data warehouse. The concept of the data warehouse is simple: extract data from one or more of the organization’s databases and load it into the data warehouse (which is itself another database) for storage and analysis. However, the execution of this concept is not that simple. A data warehouse should be designed so that it meets the following criteria:

• It uses non-operational data. This means that the data warehouse is using a copy of data from the active databases that the company uses in its day-to-day operations, so the data warehouse must pull data from the existing databases on a regular, scheduled basis.
• The data is time-variant. This means that whenever data is loaded into the data warehouse, it receives a time stamp, which allows for comparisons between different time periods.
• The data is standardized. Because the data in a data warehouse usually comes from several different sources, it is possible that the data does not use the same definitions or units. For example, each database uses its own format for dates (e.g., mm/dd/yy, or dd/mm/yy, or yy/mm/dd, etc.). In order for the data warehouse to match up dates, a standard date format would have to be agreed upon and all data loaded into the data warehouse would have to be converted to use this standard format. This process is called extraction-transformation-load (ETL).

There are two primary schools of thought when designing a data warehouse: bottom-up and top-down. The bottom-up approach starts by creating small data warehouses, called data marts, to solve specific business problems. As these data marts are created, they can be combined into a larger data warehouse. The topdown approach suggests that we should start by creating an enterprise- wide data warehouse and then, as specific business needs are identified, create smaller data marts from the data warehouse.

Benefits of Data Warehouses

Organizations find data warehouses quite beneficial for a number of reasons:

• The process of developing a data warehouse forces an organization to better understand the data that it is currently collecting and, equally important, what data is not being collected.
• A data warehouse provides a centralized view of all data being collected across the enterprise and provides a means for determining data that is inconsistent.
• Once all data is identified as consistent, an organization can generate “one version of the truth”. This is important when the company wants to report consistent statistics about itself, such as revenue or number of employees.
• By having a data warehouse, snapshots of data can be taken over time. This creates a historical record of data, which allows for an analysis of trends.
• A data warehouse provides tools to combine data, which can provide new information and analysis.

Data Mining and Machine Learning

Data mining is the process of analyzing data to find previously unknown and interesting trends, patterns, and associations in order to make decisions. Generally, data mining is accomplished through automated means against extremely large data sets, such as a data warehouse. Some examples of data mining include:

• An analysis of sales from a large grocery chain might determine that milk is purchased more frequently the day after it rains in cities with a population of less than 50,000.
• A bank may find that loan applicants whose bank accounts show particular deposit and withdrawal patterns are not good credit risks.
• A baseball team may find that collegiate baseball players with specific statistics in hitting, pitching, and fielding make for more successful major league players.

One data mining method that an organization can use to do these analyses is called machine learning. Machine learning is used to analyze data and build models without being explicitly programmed to do so. Two primary branches of machine learning exist: supervised learning and unsupervised learning. Supervised learning occurs when an organization has data about past activity that has occurred and wants to replicate it. For example, if they want to create a new marketing campaign for a particular product line, they may look at data from past marketing campaigns to see which of their consumers responded most favorably. Once the analysis is done, a machine learning model is created that can be used to identify these new customers. It is called “supervised” learning because we are directing (supervising) the analysis towards a result (in our example: consumers who respond favorably). Supervised learning techniques include analyses such as decision trees, neural networks, classifiers, and logistic regression. Unsupervised learning occurs when an organization has data and wants to understand the relationship(s) between different data points. For example, if a retailer wants to understand purchasing patterns of its customers, an unsupervised learning model can be developed to find out which products are most often purchased together or how to group their customers by purchase history. Is it called “unsupervised” learning because no specific outcome is expected? Unsupervised learning techniques include clustering and association rules.

Privacy Concerns

The increasing power of data mining has caused concerns for many, especially in the area of privacy. In today’s digital world, it is becoming easier than ever to take data from disparate sources and combine them to do new forms of analysis. In fact, a whole industry has sprung up around this technology: data brokers. These firms combine publicly accessible data with information obtained from the government and other sources to create vast warehouses of data about people and companies that they can then sell. This subject will be covered in much more detail in a later chapter – the chapter on the ethical concerns of information systems.