Three Aspects of Agile BI

The meaning of BI is fundamentally the ability of gathering information in a particular way or form in order to make better business decisions. Agile BI provides the right information to the right person at the right time. The agility comes from accessing the best data and being able to rapidly provide the correct information to help answer the question that is important for decision making. It’s not just a technology or methodology, but the combination of all: Agile BI = Self Service BI+ Agile Development method + Agile architecture

1.   Self-Service BI

Traditional BI used to be seen as being in the realm of IT because it was with IT tools that people are able to gather and manipulate information best. However, this led to centralized administration of whatever BI systems/tools are used. This centralization usually occurs within the IT department. With Agile BI, the concept is to place the administration of BI in the hands of the people who know and understand the data the best…the end business user. Agile BI is the ability for the users to quickly and effectively create their own reports from their live data and adapt to rapidly changing requirements without loss of speed or accuracy. 

• Agile BI is a BI that can follow the business requirements. In the core of a BI system are well designed, stable and reliable Cubes of data. An BI system is agile when these data can be used to merge, compare and manipulate with other data like business scenarios, trends, etc. Only then a BI system really "comes to live". And where BI should be going, and does indeed to be heading, is toward providing the data that end users can manipulate and create their own results as their business needs require. 

• Agility in BI is more concerned with how rapidly BI implementation can change to meet new business needs. The research has shown that the time window business managers have to make decisions after business events occur is being continually compressed. Often, these managers also need to see information that the BI team didn't plan for in the first place. So Agile BI is the placing of more direct control of the data into the hands of the business user, with the IT organizations enabling them. 

• More organizations turning towards a self-service mode of BI, in order to make decisions in the timeframe required. That is, corporate IT provides a robust foundation of data, the right tools, the right training, and the right support. Business managers are then operating as self-sufficiently as possible, performing their own analytics, drill down etc. The truth is, decision-windows are being compressed so much that there simply isn't time to get corporate IT involved when the "standard" BI deliverables don't provide the information required. 

• A truly agile BI solution is one that puts the ability to develop reports and dashboards in the hands of the users. But there is a gap between what we really need and what we think we need. The Agile BI concept says, “Let’s give users the capability to view their data in whatever way they’d like in order to help them uncover business opportunities that may not have been so obvious before. Give them the ability to alter reports, charts and graphs. Allow them to create ‘what if’ scenarios in order to help them decide on their next strategy.” Agile BI means self-service, better visualization, light-weight tool to enable business users to do analytics effectively. 

2.   Agile Development Methodology

Agile BI and Agile Methodology, two very different things, but intertwined with each other at the same time.  The other part of Agile BI is applying agile project delivery methodology to BI development via iterative communication & better customer satisfaction; make the tool more nimble and user-friendly.

• Agile embraces change. It is an excellent methodology for eliciting requirements when the user isn't sure what's required. By taking a pragmatic, iterative approach, the developer can produce a first cut of the solution and get it in front of a user for review without having to go through a major requirements gathering and documentation exercise . "Scrum" Project that uses smaller "iterations" of development with intensive customer involvement, and this, by implication is juxtaposed against the traditional formal requirements process driven "waterfall" project management methodology. 

• Agile BI is the implementation of the agile development (delivery) method which is effective and pragmatic. The core concept of the method is incremental iterations which is to cut the long development cycle of BI project in series of small iterations or cycles. In practice this means a close relationship between the development team and the requirement / test team(s), instead of delivering the entire project at once, the method will deliver incremental versions of the project progressively to achieve the delivery of all required functionalities.Practice 'Agile Manifesto'
  

  Individuals and interactions over processes and tools

  Working software over comprehensive documentation

  Customer collaboration over contract negotiation

  Responding to change over following a plan

• There are some infrastructure building blocks that MUST be there before you can sprint through the development. Agile only works if there is substantial work done up front to enable BI. Rapid Agile development builds on a lot of hard foundational work that comes before. THEN you can rapidly develop reporting and analytics output that works. Focusing on the agile methodology *alone* often results in:

          - A lousy foundation with decreasing velocity as changes are applied

          - A hand-over to a support team and IT governance process that seems intent on strangling the new
            agile BI thing to irrelevant

3.   Agile Architecture & Design

Agile BI is not a technology, a process, or a project methodology but a combination of all of these in addition to analytics and anything else that enables a person or organization to gather information in a better fashion for better decision making.  Defining a true BI Solution - creating successful BI strategies, processes, and applications... Know where you are before undertaking new direction-BI agility is achievable

• Agile BI is also about how to cultivate the analytics culture, break down the data silo, to shape an Agile organization as the destination (no matter which road be taken, there's should be an inspiring destination). Some part of the overall Agile BI answer is to simplify of Analytic tools/apps for easier consumption by business users. There is a very heavy technology stack that is getting in between users, their data, & effective decision making. Like methods, standards, processes (including a well-designed development lifecycle, agile or otherwise). And to a large extent, well-designed data and tools. 

• Agile BI, Agile enterprise... implies a broader decision making capability among many distributed decision makers by pushing access to the right information down and across the organization to the right end-using decision maker, on demand--hence making the organization more nimble in its decision making and its value creating market actions. The organization can instill the data-driven decision-making and analytical culture, to improve the operation and customer satisfaction 

• Agile is a mindset first and a methodology second. As a mindset, it has to permeate all areas of BI that lay between a user and a result..... There is a large overlap between structure, meaning and insight that both need to collaborate on. In that overlap we need Agile individuals who are either business savvy, IT professionals or IT savvy business professionals. The agile things include:
  1). Agile method (most discussed)
  2). Agile technology (e.g. in-memory)
  3). Agile architecture and design
  4). Agile talent  

An organization may not have an agile BI or decision making capability without  these aspects all being agile. Agile is mindset, Agile is Mythology & Technology, Agile is architecture & design, and last but not least, it’s agile talent, if the project fails, It's NEVER a Technical Problem, it's ALWAYS a People Problem

Posted in: Agile,Business Intelligence