Business Intelligence (BI)
Business Intelligence (BI) refers to the methods and tools that collect, prepare and surface operational company data as metrics, reports and dashboards, enabling fact-based decisions.
Business Intelligence (BI) is the umbrella term for the methods, processes and software a company uses to systematically collect, integrate, prepare and turn its operational data into decision-relevant information – typically in the form of metrics, reports, dashboards and analyses. The goal is to turn raw data from inventory management, accounting, sales or the online shop into reliable knowledge that managers and departments can use for fact-based decisions.
BI answers mainly retrospective and descriptive questions: How has revenue developed per channel? Which items turn over slowly? What is the gross margin per customer segment? Unlike ad-hoc one-off analyses, a BI solution delivers these answers repeatably, consistently and consolidated across data sources – as a "single source of truth" that everyone involved can rely on.
At a glance
- Turns operational raw data into metrics, reports and dashboards
- Covers data integration (ETL), storage (data warehouse) and analysis (OLAP, reporting)
- Goal: fact-based instead of gut-driven decisions
- Answers mainly "What happened?" and "Why?" – retrospective/descriptive
- Often draws its data directly from the ERP system
How does Business Intelligence (BI) work?
Business Intelligence follows a typical processing chain. It starts with the source systems – ERP, CRM, online shop, warehouse or financial accounting – where the operational data originates. This data is extracted, cleansed, standardised and loaded into a central store. Analysis tools then access it from there to aggregate and visualise it. Only this separation of operational activity and analysis makes complex evaluations possible without burdening the production systems. How current the metrics are depends on the refresh cycle: many BI pipelines load their data in nightly batch runs, while modern architectures update it in near real time.
The building blocks of a BI architecture
A classic BI architecture consists of several layers. An ETL process (Extract, Transform, Load) pulls data from the source systems, transforms it and loads it into a data warehouse – a historised database optimised for analysis. Analytical models sit on top of this data set, often as OLAP cubes (Online Analytical Processing), which allow metrics to be evaluated flexibly along dimensions such as time, region, item or customer. The top layer is formed by reporting and dashboard tools that present results as reports, KPI tiles or interactive visualisations.
Metrics and dimensions
At the core of every BI analysis are metrics (KPIs) – measurable values such as revenue, contribution margin, inventory turnover or order lead time. These metrics are broken down and filtered along dimensions, so that revenue can be shown per sales channel, month and product group, for example. A user can drill down from the total figure to the individual line item. For everyone to see the same numbers, metric definitions must be stored centrally and consistently.
Why Business Intelligence (BI) matters
The value of BI lies in decision quality. Without systematic analysis, decisions often rest on gut feeling, outdated Excel lists or contradictory individual reports. BI creates a consistent, up-to-date factual basis: sales spots low-margin customers, purchasing identifies slow-moving items, and management sees deviations from plan – all based on the same agreed figures.
A second advantage is time. Where analyses used to be compiled manually from several systems, an automated dashboard delivers the metrics at the push of a button and in near real time. This significantly shortens the path from question to answer and makes reporting reproducible rather than person-dependent. Modern self-service BI tools also allow departments to build their own analyses without IT support, provided that data models and metric definitions are centrally defined. The prerequisite for this is high data quality – incorrect or inconsistent master data inevitably leads to wrong metrics.
Business Intelligence in the ERP system
For most mid-sized companies the ERP system is the most important data source for BI, because that is where orders, invoices, stock, purchases and master data come together. In principle there are two ways to connect BI and ERP: integrated reporting within the ERP, or a standalone BI platform that draws data from the ERP via interfaces.
Many modern ERP systems already come with built-in reporting and dashboard functions that deliver standard metrics directly from the live system – sufficient for many everyday questions. However, as soon as several sources (such as ERP, shop and marketplace data) need to be merged, large histories analysed or complex models calculated, a dedicated BI solution with its own data warehouse is worthwhile. It relieves the operational ERP and enables analyses across system boundaries. The connection is usually made via an API or an ETL connector.
Distinctions: BI vs. reporting, analytics and big data
The terms around data analysis overlap but mean different things. Reporting is a part of BI and refers to the standardised, usually periodic presentation of metrics in fixed reports. BI goes further and covers the entire chain from data integration through modelling to interactive analysis.
BI, business analytics and data science
Classic BI is predominantly descriptive – it describes what happened and why. Business analytics and data science go further and ask predictive ("What will happen?") and prescriptive questions ("What should we do?") using statistical models and machine learning. In practice the boundaries blur: many BI platforms today include forecasting functions. "Big data", in turn, describes not a method but particularly large, fast-growing or unstructured volumes of data that require special technology – BI can build on such data but is not limited to it.
BI in the DACH mid-market
In German-speaking mid-sized businesses, BI has long ceased to be a topic only for large corporations. Cloud-based tools integrated into ERP systems have significantly lowered the barrier to entry, so that even smaller trading and manufacturing companies use structured dashboards. Typical use cases are revenue and margin analyses, inventory metrics, and controlling by cost centre and cost object.
Data protection must be observed: as soon as personal data – such as customer or employee data – flows into analyses, the requirements of the GDPR apply, in particular purpose limitation and data minimisation. If BI analyses are used to monitor the behaviour or performance of employees, the works council's co-determination right is also relevant. Meaningful metrics also require the underlying master and transaction data to be kept clean – without a consistent data basis, every dashboard remains deceptive.
Example
BI dashboard at a multichannel retailer
A mid-sized retailer sells through its own online shop, Amazon and brick-and-mortar stores. The revenue data is spread across the ERP, the shop system and marketplace statements – each department calculates with its own Excel lists, and the figures regularly diverge. A reliable overall view of revenue and margin per channel is missing.
With a BI solution, the data is loaded nightly from all systems into a data warehouse via an ETL process and standardised. A dashboard now shows revenue, gross profit and return rate per channel, product group and month – with drill-down to the individual item. Management thus recognises that a high-revenue marketplace channel only delivers a thin margin after returns and fees, and adjusts the assortment accordingly.
Frequently asked questions
Related services
Questions about Business Intelligence (BI) in your ERP project?
We advise vendor-neutrally – and implement it ourselves on request.