ApplauseWorld market leader for testing and digital quality assurance, has his fourth annual industry report “The State of Digital Quality in Functional Testing 2025″ Published that should help companies offer high quality apps, websites and other digital experiences. The report shows a significant increase in AI use for functional software tests that more than doubled last year – although companies continue to believe that it is absolutely essential that people are involved. Crowdtesting is an effective approach used by a third of the company to ensure comprehensive digital quality.
The users: Keep control inside when it comes to define and measure the goals of software development and quality assurance. Customer satisfaction and customer mood/feedback are the most important key figures for evaluating software quality, and tests of the user experience (UX) are still the most popular test type. However, there are still known challenges, including tight schedules and a lack of resources and stability in the internal teams. The results of the report are based on a current survey among more than 2,100 software development and test subjects worldwide.
Central findings:
AI is being integrated more and more into tests, but human control is still of the greatest importance:
- 60 percent of the respondents stated that their company use AI in the test process. Our AI survey showed that in 2024 only 30 percent used the technology to create test cases on monthly, weekly or daily basis. Ki used almost 32 percent for test reporting.
- Companies use AI to develop test cases (70 percent), to automate test scripts (55 percent) as well as to analyze test results and to recommend improvements (48 percent). Further use cases are the prioritization of test cases, autonomous test execution and adjustment, the identification of gaps in the test cover and the “self-healing” testing automation.
- AI and automation alone cannot offer the comprehensive end-to-end test cover that companies need. A third of the respondents (33 percent) uses crowdtesting as an effective approach to risk reduction through human-in-the-loop test cover, especially with increased use of agentic AI.
Despite AI efficiency, there are still considerable challenges in the pre-release testing:
- In view of the rapid increase in acceptance, 80 percent of those surveyed are faced with the lack of internal specialist knowledge in the field of AI tests.
- With 92 percent, the biggest challenge in the test area was to keep up with the rapidly changing requirements. Almost a third of the respondents are based on a test partner to close this gap.
- Further obstacles to AI quality are inconsistent/unstable environments (87 percent) and lack of time for sufficient tests (85 percent).
Companies are increasingly relying on a mixed Shift-Left approach for quality assurance (QA):
- A significant change is currently taking place in the software development life cycle. While an earlier survey showed that 42 percent of the respondents only test in a single phase of the SDLC, this year only 15 percent is limited to a single phase.
- Over half of the companies now deal with quality assurance during the planning (54 percent), development (59 percent), design (52 percent) and maintenance phase (57 percent) of the SDLC. 91 percent of those surveyed stated that their team performed various types of functional tests, including performance tests, user experience tests (UX), tests of accessibility, payment tests and more.
- Of the 83 percent of companies that use multiple metrics to monitor digital quality, 67 percent use test case reports and metrics to analyze trends and identify opportunities for improvement. 58 percent use the combined data as guidelines for future development.
“Quality assurance for software has always been a movable goal,” says Rob Mason, Chief Technology Officer, Applause. “And as our report shows, companies are increasingly relying on generative and agent AI solutions in order to drive their quality assurance measures. In order to accommodate the increasing expectations of users: at the same time and to manage the risks of the AI, it is crucial to continuously evaluate and evaluate the tools, processes and skills that we use for quality security – even before we even think of the apps And to test the requirements of performance?
Further findings:
Digital quality is customer-oriented-UX, usability and user acceptance tests as well as corresponding metrics are preferred:
- Customer satisfaction and customer mood/feedback are the most important key figures for evaluating software quality.
- User Experience (UX) testing is the most popular test type at 68 percent. This test form uses qualitative research to ensure that digital experiences are intuitive, convincing and appealing.
- Usability tests (59 percent) that measure user-friendliness and user Acceptance testing (UAT, 54 percent) are also widespread.
“Internal QA structure and consistency” is highly rated by the respondents -nevertheless there is a lack of extensive documentation.
- 69 percent of the respondents assigned the structure and consistency of their company to the framework categories “Excellence” and “expansion” in terms of digital quality.
- Nevertheless, only 33 percent stated that there were extensive documentation for test cases and plans.
- 84 percent of those surveyed find it difficult to reproduce errors based on the available test data – however, the reproduction of errors is crucial for understanding, analysis and the remedial of problems.
“The fact is that what we have predicted for a long time has now become a reality – machines can develop and validate software to a certain degree,” drove Mason Fort. “But even agent Ki requires human intervention to avoid quality problems that can cause serious damage given the speed and scope with which agents can cause serious damage. The art is to integrate human influence and security precautions early and throughout the development without slowing down the process. AI take a pioneering role. “
The State of Digital Quality contribution series From Applause offers insights into current practices and trends in the most software most and in quality assurance including preferred methods and tools as well as common challenges with which experts from software development and test are confronted worldwide. You can find more resources here:
REPORT: The 2025 State of Digital Quality
BLOG POST: Highlights from the 2025 State of Digital Quality Benchmark Survey
QUIZ: Test Maturity Quiz
About applause
Applause is the world market leader for testing and digital quality assurance. Brands win or lose customers today through digital interactions, and only applause can provide authentic feedback about the quality of digital assets and experiences, which is provided by real users in real environments. Our groundbreaking approach takes advantage of the performance of the applause platform and uses a certified community of more than one million digital experts worldwide. In contrast to conventional test methods (including laboratory tests and offshoring), we react with the speed, the scope and flexibility that digitally oriented brands need and expect. Applause delivers revealing, implementable test results that flow directly into the decision “Go/No Go” and help the development teams to develop better and faster and to publish them with confidence. Thousands of digital -first brands – including Ford, Google, Western Union and Dow Jones – Trust on Applause as Best Practice, for the digital experience