IBM’s digitalization advice: Go ‘all in’ on a few high-value processes

Modernizing legacy information technology (IT) systems begins with a plan that prioritizes high-value processes, according to an IBM executive.

“Pick four or five processes today that’s critical … and go all in,” said Dinesh Nirmal, general manager of Data, AI, and Automation at IBM. “Don’t digitalize everything in one day,” he added, “it’s not going to happen.”  

He also dismissed the mistaken notion that legacy systems are “old and crumbly.”  

“Legacy is what I call systems that run seamlessly at a much lower cost and deliver productivity, stability, and resilience day and night,” he said. 

The challenge is to take the data from these legacy systems and utilize them to serve the end-user — whether it’s a government agency or an individual consumer.  

“When you move to any new system, you have to make sure the availability of the system for the end-user doesn’t go down,” said Mr. Nirmal.  

“What do we have to do to serve our customers faster and in an available manner?” he asked. “I would say that involves taking data [in a secure manner], making it available to applications, and then modernizing those applications.”  

Several government agencies have legacy systems that have yet to be modernized. The Bureau of Internal Revenue (BIR) said that it will “rightsize” some of its offices as it ramps up its digitalization efforts; the National Statistics Office, meanwhile, said that multiple but siloed records hinder interoperability among agencies. The local government of Quezon City launched in 2021 a unified QCitizen ID to streamline its existing databases.   

“I think it [digitalization] becomes a very confusing/challenging task if you don’t have a single source of truth, because a lot of data is floating around — whether structured, semi-structured, or unstructured,” said Mr. Nirmal.  

He added that without data governance, organizations will not know how to set policies on data and ensure security.  

“This isn’t just a tech matter because it involves massive amounts of sensitive data,” he said. “It has to be done … to make sure the right set of skills are being implemented, and the right set of policies and rules are put in place.” — Patricia B. Mirasol

Leave a Comment