Agile Business Intelligence

Agile Business Intelligence

Minimize data import costs, leave business users the ability to "browse" data and have a presentation of data accessible even by users not involved in the process, but with very limited rights. This is the purpose of an Agile BI system.

Evolutionary offers a set of very specific tools to be able to launch a Business Intelligence initiative in Agile mode: leaving the business user (ie not a developer or a Data Scientist) the choice of dimensions and measures to explore.

The set of tools range from a graphical Query Builder to a Business Intelligence Server that brings data into an engine prepared for Big Data. Data processing will take place through the same Big Data engine or client tools to create Pivot Tables. Finally we will have a set of tools to present the data to the end user: the so-called Presentation Layer is represented by the same Platform.

Total integration with a Big Data engine

The modular structure was joined by an application log engine that can also be governed by users, which simply carries ALL metadata in a very high performance NoSQL engine (in the order of magnitude of billions of "records") and which is also a powerful aggregation engine "agnostic".

With this term we indicate the possibility of not having to define a priori the measurements and dimensions of analysis, but the fact that we can define them directly in the analytical engine.

agile business architecture
agile business architecture

The structure of a non-traditional BI system

A traditional structure passes through the creation of an important (and expensive) data transfer and maintenance system (the so-called ETL process, Extraction - Transformation and Loading) and the maintenance of data structures within the Data Warehouse. The data is then presented through a software called "Presentation Layer" which allows access via security levels

Both processes, ETL and data structure maintenance, take place through the intervention of IT professionals and programmers.

The Agile approach, on the other hand, tends to eliminate the ETL process, simply by loading the data into a Business application which will then itself become a Metadata layer, as well as acting as a real database (being able to contain different datamarts within it) .