Flawed outputs, overburdened IT teams, and stalled pilot projects: a survey of more than 12,000 IT decision-makers worldwide exposes how badly mid-market companies are struggling to implement artificial intelligence – and why system complexity often costs more than the anticipated benefits.
For years, AI investment has been presented as the key to remaining competitive. Yet the new study “The Global Cost of Complexity Report: The Mid-Market AI Complexity Trap” by software company Freshworks offers a sobering picture: for many mid-market companies, AI is currently more of a cost driver than an efficiency lever. The survey, which polled 12,021 IT decision-makers at director level and above across six countries – including Germany, the United States, and the United Kingdom – found that an average of 26 percent of AI spending is lost before any measurable business value is generated. In Germany alone, that translates to an estimated €2.7 billion per year.
Poor Outputs Burden IT Teams
The core problem is not a lack of willingness to invest, but rather the practical realities of implementation. According to the study, 86 percent of IT leaders surveyed say that managing complex AI systems has actually increased their teams’ workload – the opposite of what companies were hoping for. In Germany, 75 percent of IT leaders report that AI outputs produce noise, errors, or rework. The study refers to this as ‘AI slop’ – low-quality, inaccurate, or misleading AI-generated content that must be manually reviewed and corrected.
The finding is paradoxical: AI systems are currently creating new tasks faster than they are eliminating existing ones. Rather than focusing on strategic work, IT teams are spending a growing share of their time verifying AI outputs, coordinating tools, and resolving integration issues.
Pilot Projects Stuck in Place – and the Clock Is Ticking
The structural dilemma becomes especially clear when examining how far AI has actually penetrated business operations. Only 15 percent of mid-market companies have integrated AI into core business processes. Thirty-six percent remain in the pilot phase. At the same time, 89 percent plan to increase their AI investments over the next 12 to 24 months.
The gap between ambition and reality is also temporal: 72 percent of mid-market executives expect AI investments to pay off within eight months. Yet 55 percent of companies report that deployment alone takes between six and twelve months – consuming the entire expected payback period before a single euro of benefit can be measured. Within this window, there is a real risk that projects are cancelled before they ever deliver value.
Barriers Are Well Known – and Persistently Present
The study identifies three main reasons why pilot projects fail to reach production: system integration complexity (27 percent), lack of skilled talent (26 percent), and excessive configuration requirements (26 percent). These structural barriers are not new, but they remain stubbornly persistent – and they hit mid-market companies with thinner margins and leaner IT departments harder than large enterprises.
A growing tool landscape compounds the coordination burden. Mid-market companies use an average of 4.2 AI tools simultaneously; ten percent operate seven or more different solutions. Yet only 33 percent have a formal, consistently applied AI governance framework in place.
The Market Is Adapting
In light of these experiences, mid-market companies are shifting their procurement strategies. The focus is moving away from technically complex, highly customizable solutions toward products that integrate into existing workflows without extensive configuration. Thirty-four percent of IT decision-makers cite integration into existing workflows as their top priority for the next two to three years. Ninety percent prefer integrated solutions over ones requiring complex configuration. Fifty-four percent plan to expand AI capabilities by acquiring existing applications rather than developing them in-house.
Doug Farren, Executive Director of the National Center for the Middle Market, explains the cautious approach common among mid-market companies: businesses of this size are typically not early adopters of new technologies. They tend to wait until the ROI is clearly visible, using smaller pilots and tests in the meantime to validate practical viability and concrete benefits.
Methodology and Context
For the study, Freshworks surveyed 12,021 IT decision-makers in March 2026 at companies with at least 250 employees. The survey covered more than 9,000 mid-market businesses with up to 5,000 staff across six countries. The full report is available at freshworks.com/cost-complexity-mid-market-report-2026.
Dr. Jakob Jung is Editor-in-Chief of Security Storage and Channel Germany. He has been working in IT journalism for more than 20 years. His career includes Computer Reseller News, Heise Resale, Informationweek, Techtarget (storage and data center) and ChannelBiz. He also freelances for numerous IT publications, including Computerwoche, Channelpartner, IT-Business, Storage-Insider and ZDnet. His main topics are channel, storage, security, data center, ERP and CRM.
Contact via Mail: jakob.jung@security-storage-und-channel-germany.de