The hardest part of retiring an app? Making the call – TechTarget
Old apps can be surprisingly hard to retire. The challenge for CIOs is proving what the business can live without — and getting the right people to make the call.
An enterprise application might be expensive, outdated or largely redundant. A newer platform might offer better integration, improved governance and capabilities the old application never had. On paper, retiring the old system might seem like an easy decision.
But somebody still has to decide whether the business can actually live without that older app.
That’s where things get complicated. An application that is barely used could quietly support an essential business function, while a heavily used one might simply perpetuate an inefficient way of working. The people affected might have different ideas about what should stay and what should go.
For CIOs, determining when an application is ready for retirement means understanding what the business needs to preserve, what it can afford to lose and whether there’s enough evidence to make the call.
Sometimes retiring applications means asking teams to change how they work in exchange for broader business benefits.
OpenText offers an example. Between 2023 and 2025, the company reduced its software tool count from about 1,600 to 1,200. Its more than 8,000 developers used roughly 50 tools, which the company consolidated into one standardized toolkit.
“The goal was to rationalize and decommission systems,” Shannon Bell, OpenText’s CIO and chief digital officer at the time, told InformationWeek.
Cost savings were one benefit. Bell also pointed to improved product quality, simpler training and more consistent development practices. She emphasized that consolidation projects need a decommissioning phase so teams don’t simply continue using their old tools alongside the replacements.
Keeping familiar tools might please certain employees, but that decision means added cost and complexity.
Requiring a team to adjust and give up preferred applications can cause a stir, either because certain employees are resistant to change or because they know about an essential function that IT and senior management hadn’t fully appreciated.
Swiss pharmaceutical and diagnostics company Roche confronted related questions as it consolidated customer analytics systems across its global operations.
Joao Antunes, lead engineer of platforms and data strategy, described how Roche migrated local solutions to a common platform, decommissioned those no longer needed and added capabilities where they provided business value.
The decisions extended beyond software. Roche had to determine how much historical data belonged in the new environment.
“What are you going to achieve with 20 years of data that you can’t achieve with four or five years of data?” Antunes told CIO Dive. “You have to have that conversation.”
A replacement doesn’t have to reproduce everything an older application did. It does have to support the business functions that still matter.
The same question applies to established workflows: Are they still essential, or are they being preserved simply because that’s how people have always worked?
A replacement doesn’t have to reproduce everything an older application did. It does have to support the business functions that still matter.
There’s value in treating application replacement and retirement as separate but coordinated efforts.
Writing for Gartner in 2022, analyst Stefan Van Der Zijden recommended that separation. One project focuses on delivering the replacement. The other handles decommissioning the original application, including preserving information that must remain accessible.
The replacement effort needs to demonstrate that the necessary business functions work under realistic conditions.
As Van Der Zijden’s research emphasizes, the retirement effort must also account for what remains tied to the old system, including workflows, historical records, integrations and other dependencies.
Both do their work and report their findings back to the center.
Success in one effort doesn’t automatically mean the other is finished.
Fannie Mae offers another example of completed application retirements. During a multiyear effort to consolidate workflows and IT management on ServiceNow, the mortgage lender retired 49 legacy applications and introduced 52 new ones, according to Raghu Bellary, vice president of enterprise workflow technology, speaking at ServiceNow Knowledge 2025. Retiring applications and reducing the overall software portfolio aren’t necessarily the same thing.
An earlier example illustrates the historical-data problem. Writing in InformationWeek in 2016, Jonathan Feldman, then CIO of Asheville, North Carolina, described leaving an older application available for lookups after its replacement went live. Because employees continued using the old system, they didn’t discover that a needed data set had not been migrated until Feldman made it unavailable.
Keeping an old application around indefinitely can conceal problems just as surely as retiring it prematurely can expose them.
Retiring an application shouldn’t require endless analysis. But someone needs to establish what will happen when that tool is switched off — and who is prepared to live with the consequences.
• What still matters? Identify the business functions people actually need, including the obscure ones that might be easy to overlook. Don’t assume something has no value just because it isn’t used often.
• What can the replacement really do? Test the necessary functions under realistic conditions with the people who depend on them. The new application doesn’t have to reproduce every old feature.
• What else depends on the old system? Check integrations, workflows, reports, permissions and other applications. Some dependencies won’t be obvious from an inventory.
• What happens to the data? Decide what must be migrated, what can remain accessible through an archive and what can legitimately be disposed of. Make sure those expectations are understood.
• Who is making the call? Bring the replacement and retirement findings together. Resolve conflicting assumptions where possible and identify who has authority to accept the remaining risks.
• What happens after shutdown? Plan for disruption, establish who will respond if something was missed and determine how the business will confirm that the retirement worked.
The goal isn’t perfect certainty. It’s enough reliable evidence to make an informed decision — and a way to deal with what happens next.
Even after both assessments are complete, disagreements and uncertainties may remain.
Business users understand how an application supports their work. IT understands the technology and its dependencies. Executives must consider the broader consequences. None necessarily has the complete picture.
Someone — a designated person or coordinating group — needs to bring the findings together, resolve questions where possible and make sure the appropriate decision-makers have enough reliable information to say yes or no.
Van Der Zijden advocated appointing an “application undertaker” to oversee decommissioning activities. That role can provide consistency and accountability as an organization works through its retirement candidates.
But coordinating the evidence doesn’t necessarily mean having final authority. That depends on the application, the business functions involved and the consequences of getting it wrong.
Nor should every retirement require endless meetings.
Trinity Health has retired more than 740 applications through a multiyear program involving legal, privacy, clinical and financial stakeholders. Mike Prokic, its chief IT strategy officer, told HealthLeaders that retrieving data from departing vendors can be difficult when contract deadlines loom. The health system reports progress to its CEO monthly.
The problem isn’t limited to retiring software. Gartner analyst Neha Ralhan identified weak leadership, stakeholder resistance and poor communication among the reasons broader enterprise application initiatives struggle, including delays in decision-making.
A replacement doesn’t have to reproduce everything an older application did. It does have to support the business functions that still matter.
That makes the handoff from assessment to decision especially important. The coordinating person or group needs to identify what’s been established, what remains uncertain and who is responsible for accepting the consequences. If a concern is serious enough to prevent retirement, it needs to be addressed. Otherwise, another round of analysis could simply postpone a decision nobody wants to make.
CIOs won’t eliminate every uncertainty. They need to know what the business stands to lose, what will continue working and which remaining risks the appropriate decision-makers are prepared to accept.
And approval isn’t the end of the job. The transition needs to be planned around the people and processes affected, with a way to recognize problems after shutdown and respond if something important was missed.
Some disruption is unavoidable. The aim is to minimize it, make sure the right people make an informed decision and then verify that the business can actually live without the application.
James Alan Miller is a veteran technology editor and writer and Lead Editor for CIO News at Informa TechTarget. He directs coverage of enterprise technology strategy, AI, software, data, infrastructure and the decisions shaping how CIOs manage increasingly complex IT environments.
YOU MUST AGREE TO OUR TERMS OF USE AND PRIVACY POLICY.
Already a subscriber? Sign in with your email above.
Get the free newsletter read by industry experts
AI represents a powerful new tool for cybersecurity professionals, but the technology is not without risk. Discover what CISOs need to know when deciding how to use AI.
Keep up with the story. Subscribe to the free TechTarget CIO newsletter
Subscribe to TechTarget for top news, trends & analysis
Get the free newsletter read by industry experts
AI represents a powerful new tool for cybersecurity professionals, but the technology is not without risk. Discover what CISOs need to know when deciding how to use AI.
The free newsletter covering the top industry headlines
source
This is a newsfeed from leading technology publications. No additional editorial review has been performed before posting.
