An independent study reference written by Dr Phuc V. Nguyen. It is not official subject material — for assessment requirements always follow your subject outline and vUWS.
Business intelligence, analytics and data science
These labels overlap and are used loosely, so separate them by purpose rather than by technique. Business intelligence monitors a defined business using governed historical data and a standing set of metrics, on a fixed cadence, for a broad internal audience. Data analytics is the general activity of examining data to find structure, in any domain. Business analytics points those techniques at a particular business decision and extends past description into prediction and choice. Data science leans towards engineering, new data sources, and models that become a permanent part of a product or service.
Why it matters
Think about the question each one answers. Business intelligence answers whether the business is tracking where it said it would. Business analytics answers what to do about a particular decision. Data science answers whether something can be built that produces those answers automatically for the next few years. One person often does all three in a small organisation, and job advertisements mix the titles freely, so read the duties instead of the label.
A retailer needs the same twelve store metrics, correct and available by 7am daily, to every store manager. Which capability does this describe most directly?
Worked examples
A logistics firm runs a nightly delivery dashboard used by every depot manager. It also has an analyst who spent a month working out why one corridor keeps missing its delivery window. A vacancy is now open. Which label fits which work?
The nightly dashboard is business intelligence. It reports agreed metrics from governed data on a fixed cadence, and its job is monitoring rather than investigation. The corridor study is business analytics. It started from a decision a manager faced, used whatever data suited that question including messy driver logs, and ended in a recommendation. The vacancy should say which of the two it is, because the first needs discipline with definitions and refresh reliability, and the second needs question framing and method choice.
A subscription business asks whether it needs data science or business analytics to reduce cancellations.
If the aim is to understand why cancellations rose last quarter and decide what to change, that is business analytics, and it can finish in a fortnight with a clear recommendation. If the aim is a scoring service that runs every night, writes a risk value against every account and feeds the service team queue for the next three years, that is a data product, and it carries engineering, monitoring and ownership costs. Many organisations ask for the second when they only need the first.
Common mistakes
- ✗Business intelligence is just the old name for business analytics. They differ in purpose. Business intelligence exists to monitor a known metric set reliably, while business analytics exists to resolve a specific decision, often using data that was never in the warehouse.
- ✗Business analytics is a subset of data science. Neither contains the other. Data science emphasises building models and data products, business analytics emphasises the decision, and their methods overlap heavily.
- ✗Descriptive work is low value and predictive work is high value. A correct, trusted and timely description of current performance is often the most valuable asset an organisation holds, and any prediction built on a poor description inherits its errors.
- ✗The job title tells you what the work involves. Titles are applied inconsistently across organisations. Judge a role by the decisions it serves, the data it can reach, and who consumes its output.
Revision bullets
- •BI monitors agreed metrics from governed data on a fixed cadence
- •Business analytics serves one decision and reaches into prediction and choice
- •Data analytics is the technique set, not tied to any business decision
- •Data science leans to engineering, new data, and models inside products
- •Labels overlap; read the duties rather than the title
Quick check
A retailer needs the same twelve store metrics, correct and available by 7am daily, to every store manager. Which capability does this describe most directly?
An executive says the organisation needs data science to fix a pricing problem that a manager must decide on within a month. What is the most useful response?
Connected topics
More in What Business Analytics Is
Sources
- Luhn, H. P. "A Business Intelligence System." IBM Journal of Research and Development, 2(4), 314-319, 1958.The origin of the term business intelligence, framed as automatic dissemination of information to the people who act on it.
- Chen, Chiang & Storey (2012)Chen, H., Chiang, R. H. L., & Storey, V. C. "Business Intelligence and Analytics: From Big Data to Big Impact." MIS Quarterly, 36(4), 1165-1188, 2012.Traces how business intelligence and analytics broadened from reporting on warehoused data to text, web and network data.
- Davenport & Patil (2012)Davenport, T. H., & Patil, D. J. "Data Scientist: The Sexiest Job of the 21st Century." Harvard Business Review, 90(10), 70-76, 2012.The article that popularised the data scientist title and described the engineering-heavy work behind it.