Business intelligence systems are data systems built into an organization’s architecture. When done correctly, business intelligence systems enable data science at every step in the process. Organizations with sophisticated business intelligence systems automatically pull data from multiple sources into a single repository, such as a data lake, a data mart, or a data warehouse. Each of these repositories may act as a source of data for a data scientist and are designed to reduce the effort required for a data scientist to acquire data before the actual analysis can begin.
With all of the organization’s data automatically pulled into a single organizational repository, data scientists are free to focus on their strengths: using algorithms and machine learning to optimize processes within an organization or uncover hidden insights. The output of a data scientist’s work is then stored in the data repository. These concepts are visualized below.
Let’s look at an example case study: An organization is developing machine learning that identifies high-potential customers for their sales department. The purpose of this machine learning would be to help optimize the time of the organization’s sales staff, by focusing their efforts on customers who have been identified to have the potential for high-value sales. After sourcing the necessary data from the data lake or data warehouse, the data scientist builds out the machine learning necessary to accomplish this goal. The output of the machine learning script is a ranked list of customers, which can be filtered by industry, U.S. region, company size, length of customer relationship, etc.
But how is this information communicated to the sales department?
Often, the answer is a business intelligence tool such as Power BI or Tableau. Business intelligence tools can be connected to the database directly and published to the cloud, which means they can be refreshed by users as the underlying data are refreshed, with no manual effort. They’re interactive and allow salespeople to interact with, filter, or visualize the machine learning results in any way that makes sense to them. Business intelligence tools also allow us to put data into the hands of folks in an organization quickly—building out an application to display the results of your machine learning takes time. After the machine learning has been finalized, business intelligence dashboards can be created, published, and shared within an organization in hours or days, rather than months.
Have questions about deploying your data science using business intelligence systems? Feel free to reach out for a free consultation.
