In my area of work as a business engineer at Dataiku, people (customers, partners, networks, school friends) often ask me: what is the difference between business intelligence (BI) and data science?
In a previous job, I worked in a business intelligence environment. Today, I assist customers in building their own data science applications for BI purposes. So what’s the difference between these two data-centric disciplines?
What is Business Intelligence used For?
The first step to any form of business intelligence consists in gathering raw data. Once the data is gathered, data engineers use what is called an ETL (Extract, Transform, Load) tool to manipulate, transform, and classify the data in a structured database.
These structured databases are frequently called data warehouses or data marts. Typically, data warehouses are supposed to be where business owners and decision makers can access their company data and find data-driven answers to their business problems.
Thanks to modern data visualization technologies, business analysts build summaries of the data on visual dashboards, making such information accessible to a greater number of people. This information helps business users analyze past performances and adapt future strategy in light of a specific goal. What does the data say about my latest sales performances, and how can I improve them? Is the data revealing increased ROI from my advertising investments?
Business intelligence is also essential for reporting and calculating key performance indicators (KPIs). Whether top level managers use KPIs to help drive company strategy or to deliver results to shareholders, investors, and the public, business intelligence is essential to a company’s success.
When it all comes down to it, classic business intelligence provides a global descriptive vision of an enterprise's activity based on past data.
Use cases for Data Science
Data science is a discipline that involves a set of techniques and methodologies to build business applications from various sources of structured or unstructured data. It's a profession that requires technical, mathematical, and business skills.
Previously, we spoke about data storage systems such as data warehouses and data marts. In data science, another term for data storage systems is the data lake. The data lake’s purpose is to store several sources of information without aggregating operations.
Companies need people with data science skills to obtain knowledge from their data (I’ve purposefully avoided using the term big data in this sentence because you do not necessarily need big data to practice data science). The diversity of the business applications impacted by the use of data science is very large.
Here are some examples of use cases we have already worked on at Dataiku and their outcomes:
- Marketers can work more efficiently on customers loyalty
- Retailers, E-commerce, and CPG players are able to raise their conversion rates and predict their future sales by analyzing customers' behavior
- The supply chain can optimize their stocks and deliver times
- Energy and Utilities providers adapt their production to the predicted demand
- Banks and financial actors predict risks and detect frauds
- Insurance companies bid the right offer to the right customer at the right life moment
- Telco or similar subscription businesses who want to work on churn prevention
- B2C companies evaluate customer lifetime value to focus their best efforts
- Industries prevent breakdowns before they happen
- Sentiment analysis for brands by gathering Facebook and Twitter conversations
- ...and so much more!
At Dataiku, we are pleased to discover more and more use cases and ideas for howdata sciencecan be used by our users and customers on a daily basis to bring about customer experience improvements.
Some of our customers also use Dataiku to improve their BI and reporting capacities. This goes to show that these two disciplines are complementary and essential for companies who want to focus on business improvements using their data.