Agile business intelligence (BI) methodology is experiencing a resurgence as the preferred way of implementing BI programs to deliver the organisational agility and flexibility required in today’s competitive environment.
“Agile BI is not just an esoteric software development term. It is essential in creating an agile organisation,” says Michael de Andrade, MD of data management and BI specialist, EnterpriseWorx.
“The fact that the theme of the global Data Warehousing Institute focuses on ‘Creating an Agile BI Environment’ is evidence of this. So is a recent Forrester Research report, which indicates that ballooning data volumes combined with complex IT environments have put pressure on legacy BI tools.
“Multiple components including data integration, cleansing, modelling, warehousing, metrics creation and management reports must be fitted together in a meaningful BI application,” says de Andrade. “But stakeholders are largely dissatisfied with BI applications’ agility and flexibility.
“Agile BI must be considered as a way to get development and implementation done faster and to react more quickly to changing business requirements. It calls for incrementally delivering products versus detailed planning and specifications.
“BI affects the entire business. At the outset, it’s important to find out from executives what key performance indicators they need to run the business so as to develop appropriate charts and dashboards. This involves carrying out a range of interviews. Using an agile process, we can gather and analyse the requirements at the same time, and start delivering on the first requirements, while gathering others.”
BI is an iterative process not a one-stop destination. It is open to ongoing fine-tuning that allows management to obtain more and more insights from their information assets, so as to create an agile organisation.
Some organisations value consistency first and foremost, and focus on putting rigid structures, processes, data standards and data governance processes in place prior to embarking on any IT project, including BI. Others perceive such processes and standards as too restrictive, cumbersome and slow, believing it’s more important to be nimble and adaptable.
In de Andrade’s view, what is required is a balance. “Organisations need to put appropriate standards and processes in place that can support speed, change and adaptability – and also consistency,” he says.
EnterpriseWorx ensures that projects are delivered on time and within budget by using agile project management methodologies such as Scrum. This is a tried-and-tested iterative development framework, which contains a set of practices and predefined roles, used to manage complex projects. It is ideal for any project involving database and BI development because it allows the management team to quickly identify and resolve bottlenecks.
“This means that customers get their software customisation and implementation projects in usable phases, which speeds up delivery and provides business benefits more quickly,” says de Andrade.
“As with any IT project, it is important to draw up a project definition document that states the expected outcome and includes an assessment of the risks inherent in implementing the system, along with ways of mitigating these.
“Forward planning is the key to being agile in practice. However, although ‘agile BI’ as a methodology can assist in creating organisational agility, it is not the same thing. Constant change has become the norm, and organisations need to embrace agility as a business imperative if they are to succeed in a world that moves at breakneck speed.”