The new paradigm of Business Intelligence

Traditional Business Intelligence (BI) solutions have been improving decision making for industries for quite some time now. This holds true especially with the growth of business analytics, both from the analysis and visualization point of view.

Let us ask an honest question to ourselves: “Is this is enough in today’s business context across industries?” The answer is that it may be helpful, but only to a certain extent.

The traditional enterprise data warehouse can store transactional information captured from multiple systems, including Excel/flat files that have some structure or pattern. An analytical model can be created with or without an OLAP layer to provide rich analytics in terms of dashboards and reports. Mining models can be created and integrated with a data warehouse or OLAP layer to build predictive analytics. At the end, these business analytics tools should help the target audience to find out ‘What’, ‘When’, ‘How’, ‘Where’ and ‘What may (will)’ happen with respect to their business.

The Independent Software Vendors (ISVs) belonging to various industries have built their products and solutions using business intelligence modules to provide better insights of the business processes. However, these business analytics are limited to the information that has been captured through the structured software.

Today, we have greatly evolved in terms of capturing vast amounts of non-transactional, unstructured business data from disparate sources and storing it. This has inevitably paved the way for a new generation of business intelligence tools and such as Power BI. The usage of ‘Hadoop’ as a Big Data platform either in addition to enterprise data warehouse or as substitute has started to rise. The key drivers for this paradigm shift are:

1. New Business Insights (viz. The Internet of Things)
2. New Technologies (viz. AI ML and data engineering)
3. Reduced Costs

The new technologies help in enhanced data management, new deployment options, and advanced analytics for ISVs to make their software products suitable for today’s business needs. This is the game changer for ISVs. Usually, in the healthcare, retail and e-commerce space, the applications have progressed by leaps and bounds. Here are some top use cases that are applicable for ISVs across industries:

1. Optimize Funnel Conversion by achieving more growth in sales of product/service with lowered costs
2. Behavioral Analytics to help analyze customer/employee behavior and customize offerings to maximize gains
3. Customer segmentation for improved targeting of the right products/services at the right place and the right time. This will be more accurate than first generation BI as the results will be in combination with macro environmental non-transactional data associated with the transactional data captured through traditional source systems
4. Predictive analytics for better planning and forecasting. The results of predictive models are more accurate when we feed a larger volume and a variety of data in contrast to the first generation of business intelligence technologies
5. Market Basket Analysis to better bundle the service offerings to the right target audience
6. Predict security threats by analyzing historical breaches and prepare better for the future
7. Fraud detection to identify possible fraudulent activity beforehand and minimize losses

In this new era of social, mobile, analytics and cloud (SMAC), the data is generated through either sensor systems, machines, mobiles, web logs or interactions through social media. This data is unstructured, voluminous, and varied. However, it is critical when combined & analyzed with the structured data captured through traditional systems. Today, after a transformation over the course of time, the new eco system of business intelligence includes two critical components:

Data Refinery

• Ingests raw detailed data in batches and/or real-time data into a managed data store
• Distils the data into useful information and distributes results to other systems
• Key use of Hadoop today

Computing Platform

• Used for exploring data and developing new analytics models
• May also be used for prototyping new analytics-driven LOB applications and for temporary analytic solutions
• May employ an RDBMS or Hadoop
• Enables users to blend new types of data with existing information to discover ways of improving business processes
• Allows users to experiment with different types of data and analytics before committing to a particular solution
• May employ a RDBMS or Hadoop-based solution running on premises or in the cloud – Hadoop is especially well-suited to processing large amounts of multi-structured data
• Represents a shift in the way organizations build analytic solutions:
  • Increases flexibility and provides faster time to value because data does not have to be modelled or integrated into an EDW before it can be analyzed
  • Extends traditional business decision making with solutions that increase the use and business value of analytics throughout the enterprise

This way the paradigm shift has taken place in business intelligence tools and platforms. Unsurprisingly, it is also creating a dire need for ISVs to modernize and build their software products and applications with above capabilities. The ISVs that have sensed this and have started implementing the change are gaining a competitive advantage.

The rise of Cloud and Disrupting the BI / DW ecosystem

Business Intelligence analytics tools and BI analytics are ushering in a new age of business. Combined with mobile BI and open-source BI tools, the BI market will surely surge ahead.

The ISVs have already realized the potential of cloud, and moved their software and applications there by offering SaaS based services to the customers. The next big question in their mind is: “Can a Business Intelligence module can be moved there?” This has its own advantages and apparent challenges that can be addressed. This has given rise to ‘AaaS – Analytics As a Service’. Stay tuned for our next blog where we deep dive into this..