Poor Quality Software Costing UK Public Sector Millions – New Report

Research from software testing firm Tricentis indicates that poor quality software is hitting the UK public sector hard. It claims that on average it drains £1.34 million a year, with 40 percent of organisations losing more than £773,000 annually.

The risks extend far beyond financial loss: outages and system failures undermine both citizen-facing services and internal employee systems. Nearly two-thirds (62 percent) of public sector organisations predict that they are at risk of an outage in the next 12 months, and more than a third report falling team morale and higher staff turnover because of poor software quality.  

The findings of the 2025 Quality Transformation Report show that software quality issues are not just an IT concern but a barrier to delivering stable, secure, and citizen-focused services.

Key Findings From UK Public Sector

Public sector software quality carries a high price tag: the average cost of poor-quality software totals £1.34m per year. Four in 10 UK public sector respondents report that software quality issues cost their organisation over £773,000 annually.  

Sixty-two percent of UK public sector respondents said they are at risk of a software outage or impact in the next year – more than manufacturing businesses (60 percent) and close behind the retail industry (64 percent).  

Nearly a third (31 percent) highlighted increased likelihood of more frequent security breaches or compliance failures because of poor quality software, which carry both financial and reputational consequences.  

Elsewhere, poor quality software is leading to delays. One in five public sector decision-makers said they had full confidence to release software on schedule. 

Additionally, more than a third (36 percent) of public sector-based respondents identified lower team morale and higher staff turnover as a key result of poor-quality software. 

The top barriers to improving software quality include the pressure to release software too quickly (43 percent), lack of personnel to handle the work (31 percent), lack of skill (27 percent) and inadequate communication between developers and testers (27 percent). 

Meanwhile, 77 percent of UK public sector organisations said they routinely release untested code, creating vulnerabilities and performance issues. 

“The gap between AI confidence and necessary safeguards raises key accountability questions. When benefit payments are delayed or NHS systems fail, public trust in essential services is undermined. Ensuring digital services remain stable and reliable is one of the biggest challenges for public sector departments today,” said Andrew Power, head of UK and Ireland, Tricentis.

Lack of AI Guardrails

The report also revealed that nearly all (92 percent) of UK public sector technology leaders plan to increase AI usage in software quality assurance over the next year, and 88 percent are confident in autonomous AI making software release decisions.

However, the research also highlights that important guardrails are missing. Only 34 percent of public sector respondents feel they have sufficient controls in place to validate AI-driven decisions, while 12 percent believe their organisation does not have adequate safeguards.

“As AI becomes more embedded in government operations, the margin for error narrows. We need to hold public sector software to the highest possible standards to protect taxpayer funds, maintain public trust, and ensure that digital transformation delivers real-world benefits,”said Power. 

Tricentis questioned 2,750 respondents from 10 countries and five industry verticals. 500 respondents were from the UK, 77 of which were from the public sector.