Skip to main content Skip to Footer


Federal IT: Modernize with confidence

Two ways to start smart


Rising maintenance costs. Growing security gaps and threats. A changing workforce. Opportunities to drive greater agility and responsiveness to changing needs. There’s little debate about why federal agencies need to modernize their IT infrastructures, including their legacy applications. But in the face of large, high-profile failures, there’s also plenty of apprehension around how best to approach application modernization (AppMod).

Building a business case and creating a roadmap are both important—and logical places to start. Yet neither addresses one of the biggest challenges to AppMod success: shining a light into the “black box” of legacy applications.

Take an X-ray

Talk to people who have tackled an AppMod initiative and you’re likely to hear war stories about all kinds of surprises: Systems believed to have 15 interfaces that actually had 35. Or applications assumed to be written only in COBOL that were harboring additional code written in Assembler or other legacy languages. Such surprises are costly and risky. As it turns out, they’re also unnecessary.

With the right tools and skills, agencies can take a deep dive into applications and decide whether, how and when to modernize them. This process of “application discovery” is a crucial prerequisite to AppMod success.

A strong application discovery toolkit makes it possible to take an “X-ray” of virtually any legacy application. The discovery process should help map application functionality—including business processes, business rules, points of integration, code complexity and sources of data. When such a toolkit has automation-assisted capabilities, it can produce a blueprint containing those insights in as little as a few weeks. Such blueprints are invaluable in guiding AppMod decisions, supporting cost estimates and, most importantly, eliminating surprises.

Try before you "buy"

Application discovery isn’t the only way to increase the odds of AppMod success. Another option is a proof of concept (POC). A chance to kick the proverbial tires, a POC helps to mitigate risk and boost confidence in everything from project estimates and costs to vendor selection and scope definition.

Working with us, federal agencies are realizing the value of POCs firsthand. One agency successfully converted more than 35,000 lines of COBOL code into Java and migrated data from legacy Datacom files to an Oracle database—all as part of a 10-week POC. For another federal agency, we spent just eight weeks converting more than 30,000 lines of COBOL into a Java-based application. In these and many other cases, agencies demonstrated like-for-like functionality. They confirmed the ability to host in the cloud. And, above all, they were able to move forward with a go or no-go decision based on actual experience—not hopes or fears.

Proof Of Concept:
A formal test of a proposed approach to modernizing a single system or a portion of code from a system

Make the tackle

Upfront planning and analysis followed by incremental execution to mitigate business and technical risks will avoid costly surprises and accelerate success. How to start? Build a business case. Create a roadmap for how apps will be modernized. Use application discovery to identify and mitigate risk at the start. Then try before you commit—validating proposed strategies with one or more proofs of concept.

When it comes to AppMod, a smart start leads to a strong finish.

The challenge of modernization

Technology in Transition: IT Modernization

As government systems age, security risk goes up and cost efficiency goes down.


Eric Stogoski

Eric Stogoski
Senior Manager

Connect with Eric Stogoski's Profile on LinkedIn. This opens a new window.

Mail to Eric Stogoski. This opens a new window.


Stay In The Know

Receive e-mails from Accenture featuring new content that matches your interests.

Visit the subscription center to make your selections and subscribe to New from Accenture.