What is BI Business Intelligence and to whom it is intended

Business Intelligence (BI) is technology led data analysis process and representing useful data necessary for leaders, managers and other corporate users to make positive and right business decisions. BI includes a wide spectrum of tools, applications and methodologies which enable organizations collecting data from internal information systems and/or external sources as a result of business processes (transaction databases – e.g. ATM cash flow; remote readers of electrical energy consumption, water, gas), Excel tabeles and other forms of generated reports.

BI prepares data for analysis based on created data queries, creates reports and visualizes data in order to make analytical results available to decision-makers as well as operative workers.

Reasons and needs for introducing BI solutions

• Existing of need to increase revenue, reduce costs and operate more competitive

It’s been a while when end-users could make business decisions using monthly reports. BI accelerates and improves decision making, optimizes businees processes, increases operational efficiency, influences the creation of new revenue and gains competetive advantages over competitors.

• Need to display a small amount of data, which are result of processing and analysis of large data quantities

BI data may include historical information stored in the data storage, as well new data collected from different sources that enable BI tools to support strategic and tactical decision-making processes.

• Existence of need to cut IT expences

In the begining, BI tools were primarily used by data analysts and other IT professionals, who conducted analyzes and produced reports with query results. However, business executives and workers increasively use BI platforms, partly thanks to development of self-serving BI tools.

• Existence of need for permanent tracking of company performance

Company is required to clearly define its goals and plans, and regarding to these, key performance indicators (Key Performance Indicators) are interpreted.

Architecture of BI – Business intelligence

Data sources

Basically, there are two types of data sources: external and internal.

Internal data belong to the company and become as a result of generating through a transation system, desribing activities that have happen within the company, such as: financial subsystem, sales subsystem, production, resource monitoring, billing …..

External data are obtained outside the company, most often using specialized functions that deal with the collection and distribution of data. They are of critical importance for stratigic decisions because with them company sees favorable opportunities as well as threats. External data may refer to competitiveness data (products, services changes of competitor companies….), economic data (currency fluctuations, political indicators, interest rates movement, stock data), professional ( technological trends, marketing ….), econometric (income of individual groups, customer behavior ….), psychometric (customer profiling), demographic and marketing data.

Integration services (Extraction, Transformation, Loading)

Integration services are tools that are used for wide spectrum data migration. They are used for data transfer, extraction, transformation and loading (ETL). Integration service also has data import and export wizard which can work with different types of data without transformation.

SSIS (SQL Integration service)

DW – Data Warehouse

DW presents data warehouse. Data must be integrated, consolidated and cleared using integration services before storing them into DW. Data can be stored in a smaller DM (data marts) which are grouped by business processes: finance, production, warehouse, sales ….. which represent subgroups of company information.

Each data contains a timeline with a historical track on the company’s business as well data from an external environment. Prediction of future events can not be performed without knowledge of past of the same or some other events. DWH is designed in a way to enable data searching, (OLAP) on-line analytical processing, reporting and supporting process of business decision-making.

OLAP and OLTP technology (analytical and transactional databases)

To understand the essence of the DWH concept, it is of great importance to notice basic  characteristics, advantages and disadvantages of OLTP transactions and OLAP analytical data processing.

OLTP (On-Line Transaction Processing) Oriented on details, with frequent updates from end users side, continuous (permanent) business processes. OLTP systems are based on relational databases in accordance with normalization rules. Complex information is divided into the simplest structure-tables. They are suitable for fast data update, establishing relationships, while complex reports preparation can take a lot of processor time and break database performance.

OLAP (On Line Analytical Processing) is type of technology that allows analysts and managers insight to data through fast, consistent and interactive access to large number of various reports based on information obtained through the transformation of raw data. This data is denormalized, running queries with such data organization is much faster, at the same time simplifying database schema. That way, data searching is made easier for staff without technical background. The model is based on methodology of multidimensional analysis, which means that data can be viewed through a great number of filters. These systems do not deal with data processing, but with interpretation and analysis.

OLAP vs OLTP       

Property	OLAP	OLTP
Characteristic	Information processing and data analysis	Operational data processing
Focus	Complex queries	Large number of transactions, data entry
Users	Managers, Analysts	Salesman, technician, a multitude of different users
Database size	Gb – Tb data	Mb – Gb data
Access / number of users	Read / hundreds	Read – write / thousands
Functions	Long-term analysis, summarized and processed data	Daily operations (raw data / input, editing …)
Data type	Historical data	Updated data, current
Database design	Star / snowflake	ER model

Reporting and Analytical tools

BI technology also includes visualization software for graph design and other infographics as well as tools for creating table, Dashboards that display visual information about business metrics and KPI in a simple way. Data visualization tools have become standard part of modern BI. Several leading manufacturers defined technology of BI tools for reporting and analysis.

Reporting services

BI trend

Besides BI Managers, business intelligence teams usually include a mix of BI architects, BI developers, business analysts and data management professionals. Business users are often people who represent business side, who are not directly interested in the details of DWH, but they are there to ensure business logic in the process of BI development.

BI platforms are increasingly used as front-end interfaces for large data systems. Modern BI software usually offers flexible endpoints, allowing connection to a variety of data sources. This, with a simple user interface, makes a good tool for managing a large amount of data.