Short vision on Business Intelligence (BI) and CI/CD
Business Intelligence (BI) is about taking care of business decision making by joining facts, brains and hearts in an ever more digitizing world. Screens, business logic, software and data are connected but not locked in anymore. CI/CD philosophy and Low-Code development enable organizations to better adapt with challenges, creating a safe fail environment together along the way.
I developed a BI target operating model back in 2007, trained students with it. During the following years, the model adapted to incorporate Agile influences and increased reusability and automation of software/-development as my role as consultant got divers and more mature. The model presented above, with adaptations for Delft Municipality, functions as a map to talk with non-tech people and how their job relates to successful delivery of BI products. It demonstrates the flow from demand to supply with a feedback loop directly with end-users to check delivery on set goals.
In almost 15 years, the core flow has not changed. Agile practices allowed for expert brainpicking in early stages of demand formulation, so to sharpen the view of end results and help business in their new relationship with IT. Business environments change faster, client demand evolves, legal frameworks change, political wind becomes turbulent, where to go and what to do to stay in business. What goes for BI also goes for other technology efforts: shorten the learning cycle, create a safe-fail culture instead of maintaining fail-safe practices where business demand outpaced IT supply.
A recent perspective on the world of software development, including Business Intelligence, is called Continuous Integration / Continuous Delivery. Well, I am not an expert on CI/CD yet I do have an opinion on the matter. I welcome it very much since for me it is not so new actually. In my opinion the CI/CD operating model is not so different from old skool OTAP on steroids using current automation practices. I like the First Consulting view on the build-release pipelines, presented below.
My view and hand hands-on experience with CI/CD in Business Intelligence environments seem to overlap Informatica Big Data Management tool explained in the Informatica blog here and in the Informatica whitepaper here. I used two examples by Informatica because I recently was approached for an organization that has chosen Informatica for its CI/CD activities, amongst other functionality.
Automation and code generation enable a more iterative process of working together in multi-disciplinary teams where priorities can swiftly change. I am an advocate of Low-Code software development (experienced in Outsystems), creating enterprise grade applications as if it were a PowerPoint. A modular approach to integrating legacy enables business a stable transition to new applications that better meet future demands.
Business Intelligence novel practices and low-code software development combined will prove extremely useful for incumbents, the oil tankers, to speed up their digital transformation, delivering functionality and intelligence at lightspeed. But be aware, so can startups too, the speedboats, from their little atticks, reach out to millions of customers and take a (large) portion out of your market share.
You should think about which challenge you as a corporation, would like to tackle first: internal politics hindering innovation or negotiations with unions because your business is loosing out on the competition. Better make that decision informed 🙂