What is business intelligence?
Business intelligence (BI) is the set of processes and tools an organisation uses to collect, manage and analyse its data so that decisions rest on evidence. IBM’s definition is “a set of technological processes for collecting, managing and analyzing organizational data to yield insights that inform business strategies and operations” [2]. In practice BI covers everything from a monthly sales report to a live dashboard that flags a problem in a warehouse within minutes.
BI is mainly descriptive. It tells you what has happened and what is happening: how many customers you won last month, which region is behind target, whether order sizes are going up or down. Business analytics, which IBM treats as a subset of BI, is the forward-looking part that suggests what to do next [2]. Modern BI platforms, like the one Intellicus describes in the infographic above, increasingly bundle both.
The core components of business intelligence
Whatever product you use, a BI system is built from the same parts. IBM describes the usual flow in five steps [2], and each step relies on a component:
- Data sources. The raw material: sales and finance systems, CRM, inventory, supply chain, marketing platforms, industry statistics and social media.
- Data collection and integration. Gathering and cleaning the data, by hand in a spreadsheet or automatically with an extract, transform and load (ETL) process.
- Data storage. Traditionally a data warehouse, which brings data from many sources into one central system. Data lakes and data lakehouses are now also used [2].
- Analysis. Looking for trends and unexpected results with queries, data mining and data modelling. Many warehouses include an online analytical processing (OLAP) engine for multidimensional questions such as “sales in the east versus the west this year, compared with last year” [2].
- Reporting and visualisation. Charts, reports and dashboards, ideally with drill-down so users can move from a summary to the detail behind it.
The final step is the one people forget: an action plan. BI only pays off when results are compared against key performance indicators and someone changes a process, a budget or a campaign because of what the data shows [2].
Key business intelligence concepts
A handful of ideas come up in any BI project. These are the ones worth knowing.
- Single source of truth. The point of a warehouse is that everyone reports from the same cleaned data, so finance and sales stop arguing about whose figure is right.
- Descriptive, predictive and prescriptive analytics. Descriptive analytics says what happened; predictive estimates what is likely to happen; prescriptive suggests what to do. IBM places the last two under business analytics [2].
- Drill-down. Moving from a total to the detail underneath, such as from a country to a store to a single product, to find the cause of a result.
- What-if analysis. Changing an assumption, such as a price or a budget, to see the likely effect. The Intellicus graphic lists it among the core features of modern BI.
- Key performance indicators. The handful of measures a team agrees to track, which give dashboards their purpose.
Modern BI capabilities
The infographic picks out four capabilities that BI platforms added on top of the traditional report and dashboard.
Machine learning based analytics
Machine learning looks for loosely correlated patterns in business data that a person running standard reports would miss, such as a combination of factors that predicts customer churn. The graphic claims 82% of market leaders were adopting AI and machine learning, but gives no source for the figure [1].
Predictive analytics
Predictive models use past data to estimate future outcomes, such as demand next quarter or the chance a customer will pay late. The graphic quotes a forecast that the predictive analytics market would grow by around 21% a year between 2016 and 2022 [1].
Real-time analytics
Real-time analytics processes data as it arrives, so that problems can be caught while they are still happening. Typical uses are fraud detection, monitoring production lines and tracking website or app performance. The graphic cites a survey in which 60% of CIOs at large companies said they used real-time analytics to monitor critical processes.
Self-service analytics
Self-service tools let business users build their own reports and explore data without waiting for an analyst. Gartner predicted that self-service BI platforms would make up 80% of all enterprise reporting by 2020, according to the graphic [1]. IBM notes the downside: when every team builds its own analysis, different teams can reach different conclusions from the same data [2]. Clear data definitions and governance matter more, not less, once self-service is switched on.
How old is business intelligence?
Older than most people expect. The term was first used in 1865 by the writer Richard Millar Devens, describing a banker who gathered market information before his competitors. In 1958 the IBM computer scientist Hans Peter Luhn explored how technology could be used to gather business intelligence, and his work helped shape early analytics systems [2]. The data warehouses, OLAP engines and dashboards that most organisations use today grew out of the decades that followed.
Are the infographic’s market forecasts still relevant?
Treat them as a snapshot of how the market looked around 2018, when the graphic appears to have been made. Every forecast on it ends between 2020 and 2023: a data visualisation market of USD 7.76 billion by 2023, self-service BI software growing at 15.2% a year from 2016 to 2021, and advanced analytics growing at about 15% a year from 2017 to 2023 [1]. Apart from the Gartner prediction, the graphic does not say where these figures come from, and we could not trace the original reports. The direction they describe, with more automation, more machine learning and more self-service, has held. The exact numbers should not be quoted as current.
It is also worth remembering that Intellicus sells BI software. IBM’s caution that installing BI tools does not by itself create a data-driven culture is a useful counterweight [2]: the components above only help if people trust the data and act on it.
Questions about business intelligence concepts
What are the main components of business intelligence?
Data sources, data integration (often ETL), a data warehouse or similar store, analysis tools such as OLAP and data mining, and reporting and dashboards [2].
What are the basic concepts of business intelligence?
Collecting data into one trusted store, analysing it to find trends and causes, presenting the results in reports and dashboards, and acting on them against agreed KPIs [2].
What is the difference between business intelligence and business analytics?
BI is descriptive and shows what has happened. Business analytics is forward-looking and suggests what to do next. IBM treats business analytics as a subset of BI [2].
What are examples of business intelligence tools?
Common tools include Tableau, IBM Cognos Analytics, SAP and Microsoft Excel, which IBM names as examples for building visualisations and dashboards [2].
Who first used the term business intelligence?
Richard Millar Devens, in 1865 [2].
Sources
- Intellicus, Business Intelligence, For Now and For the Future, the original infographic.
- IBM, What Is Business Intelligence (BI)?.
Last reviewed 2 November 2026
