Sergey Nivens - Fotolia
For some large enterprises, technical debt accrued over many years, whether it is through ongoing maintenance of legacy systems, mergers and acquisitions or poor software development practices, remains one of the biggest impediments to digital transformation.
According to a global study of over 500 IT decision makers, businesses devoted more than a quarter of their IT budgets to addressing technical debt compared to about a third to innovating and building new capabilities, and just shy of 40% to running status quo operations.
The problem was even more pronounced for enterprise companies, which spent roughly $4 out of every $10 in their IT budgets on technical debt, according to the study that was commissioned by OutSystems, a supplier of low code application platforms. That leaves just 30% or less in the IT budget for both status quo operations and new capabilities.
While the causes of technical debt are manifold, the top two reasons cited by respondents were the sheer number of development languages and frameworks that stood in the way of modernisation efforts, as well as turnover in development teams.
Developers could also be procrastinating on bugs that needed to be fixed to meet production deadlines. The bugs snowball over time, plunging their companies into a vicious circle of technical debt.
According to the study, which was conducted by Lucid, a research firm, 43% of respondents noted that accepting known defects to meet release deadlines was a big or critical problem.
In Australia, India and Singapore, the three Asia-Pacific (APAC) markets covered in the study, staff turnover and limited staff skills were less challenging when compared to their peers in other regions. Also, respondents in the APAC markets indicated that outdated development languages and frameworks had less of an impact compared to those in Europe, Middle East and Africa.
Read more about IT in APAC
- Australia’s Macquarie Bank has moved all its data and analytics to the cloud and is applying machine learning to detect fraud and improve customer experience.
- Singapore telco M1 is dialling up its digital chops by monetising its data, building more personalised services for subscribers, and going beyond connectivity as it evolves into a digital telco.
- Indonesia’s Bank Jabar Banten Syariah is adopting DevOps and microservices to speed up delivery of applications.
- India’s Bharat Petroleum has developed a digital nerve centre powered by the internet of things and artificial intelligence technologies to monitor the journey of its products.
Mark Weaser, vice-president of Outsystems in APAC, said organisations in the region tend to see legacy systems as a form of technical debt.
“Almost any larger company would have had technical debt, and it’s been laid bare by the pandemic as all of a sudden, IT departments are being asked to digitise very quickly and build interfaces to customers that didn't exist before,” said Weaser. “And they have a limited budget – probably in most cases a lower budget – and then they see all the money that they have to spend to maintain these legacy systems.”
Globally, nearly seven in ten companies in the study said technical debt had limited their capacity to innovate. Just over 60% of companies reported that technical debt is a significant drag on their performance, and an even higher percentage (64%) said it will have a major impact on their future business.
Given the sheer size and complexity of technical debt, respondents were also asked about their ability to address the issue now and in the future. While a small percentage of them expressed confidence in current efforts to tackle the problem, a higher number had a more upbeat outlook, which signals they may be underestimating technical debt’s tendency to compound over time.
On what is being done to address technical debt, the study found that reducing turnover and the number of development languages was a high priority, followed by addressing incorrect architectural decisions and the impact of changing governance or regulatory requirements on code.
Meanwhile, some transformation projects, such as robotic process automation (RPA), are still going on even as organisations continue to chip away at their technical debt.
In a podcast on the role of technical debt in the post-pandemic landscape, Brian Hopkins, vice-president and principal analyst at Forrester Research, said organisations do not have to retire debt for RPA to work. Citing an example of an airline that implemented RPA to automate the human effort required to process the surging number of cancellations at the peak of the pandemic, Hopkins said the company had been able to “wire around” their technical debt in just seven days.