Study: Tech Friction Drains 51 Workdays Per Year Despite Record AI Investment
6 min read
Enterprise AI investment is at record levels. Yet WalkMe's fifth annual State of Digital Adoption report — a global survey of 3,750 executives and employees across 14 countries at enterprises with 1,000 or more employees — reveals something that should fundamentally reframe how organisations think about their AI programmes: employees are not adopting the tools. More than half (54%) bypassed AI tools and completed tasks manually at least once in the past 30 days. A further 33% have not used AI at work at all. This is not friction — it is rejection. And the cost is quantifiable: workers now lose the equivalent of 51 working days per year to technology friction, up 42% from 2025, and now at a three-year high. Despite a 38% increase in digital investment from last year, 40% of that spend underperforms.
The Executive-Employee Perception Chasm
The most structurally significant finding in the WalkMe report is not the technology friction figures themselves — it is the scale of the gap between how executives and employees describe the same organisation. The survey finds what might be charitably called a parallel reality problem, and less charitably, a governance failure. Three data points define the chasm:
- AI trust: a 52-point gap — Only 9% of workers trust AI for complex, business-critical decisions. Among executives, 61% do. The 52-point differential is not a disagreement about AI's trajectory — it is a divergence about present-tense reliability that fundamentally undermines the business case for autonomous AI deployment in the absence of trust-building infrastructure.
- Tool adequacy: a 67-point gap — 88% of executives are confident that employees have adequate tools. Only 21% of workers agree. When less than a quarter of the workforce believes it has what it needs to do its job effectively, the organisation has a fundamental disconnect between technology investment decisions and technology deployment reality.
- Productivity impact: 81% vs 7.9 hours/week lost — 81% of executives believe they have significantly improved productivity through AI. Meanwhile, workers waste 7.9 hours per week dealing with digital frustrations — the equivalent of losing one full working day every week. Across a working year, this compounds to 51 days of productive capacity evaporated per employee, per year.
The trajectory of the friction figure makes the situation more alarming, not less. Technology friction had been improving: it fell from 43 days in 2024 to 36 days in 2025. The jump to 51 days in 2026 — a 42% increase in a single year — coincides directly with the rapid deployment of AI tools at scale across enterprise environments. The implication is clear and uncomfortable: AI deployment, executed without adequate adoption infrastructure, change management, and governance, is actively making the technology friction problem worse rather than better.
"The problem is not AI's capability. The technology will keep improving. What won't improve on its own is the human side: the trust gap, the governance gap, the question of who acts, when, and with what guardrails. That's what this data is really showing. And that problem doesn't go away as AI gets smarter. It gets harder."
— Dan Adika, CEO & Co-Founder, WalkMe
The Shadow AI Problem: Rejection Without Governance Creates Its Own Risk
When sanctioned AI tools fail to meet employee needs, the workforce finds alternatives — and those alternatives carry their own risk profile. The report finds that at least 45% of workers used unsanctioned AI tools in the past 30 days. Of those, 36% did so with confidential data — a data governance and security exposure that is directly attributable to the gap between what employees need to work effectively and what officially approved tools can deliver.
The organisational response to this pattern has been largely punitive in intent but structurally ineffective in execution. 78% of executives say they want to discipline shadow AI use — yet only 21% of workers have ever been warned about AI policies. A further 34% say they do not even know which AI tools their employer approves. It is difficult to enforce a policy that most of the workforce is unaware exists. And the executives themselves are not entirely convinced that enforcement is the right response: 62% of executives agree that the risk of unsanctioned shadow AI is overstated compared to the risk of not taking enough advantage of AI in the first place — a contradiction that reveals the genuine ambivalence at the leadership level about how aggressively to govern employee-led AI adoption.
"The use of shadow AI isn't a behavior to penalize; rather, it's an opportunity to address a systemic gap. When employees use unapproved AI tools, they're compensating for performance or efficiency gaps left by sanctioned tools and unclear governance. Organizations that close this gap by equipping AI with real-time context, cross-application reach, and robust guardrails will ultimately realize the strongest return on their AI investments."
— Keith Kirkpatrick, Vice President and Research Director of Enterprise Software & Digital Workflows, The Futurum Group
The Investment Paradox: More Spend, Worse Outcomes
The WalkMe report's finding that a 38% increase in digital investment has coincided with a 42% increase in technology friction and a return of productivity loss to a three-year high is a direct indictment of the prevailing enterprise AI deployment model. The dominant approach — procure more powerful tools, deploy them rapidly, measure adoption through licence utilisation or executive-reported sentiment — is systematically failing to close the gap between tool capability and employee utility. When 40% of digital investment underperforms, the question is not whether the tools are powerful enough. The question is whether organisations are investing in the adoption infrastructure — training, change management, contextual guidance, governance frameworks, feedback loops — that determines whether powerful tools become productive tools.
The friction trend reversal is particularly instructive. That organisations had achieved a reduction from 43 days of friction in 2024 to 36 days in 2025 demonstrates that improvement is achievable — it did not happen by accident, and it was not driven by tool capability alone. The subsequent jump to 51 days in 2026 happened because AI deployment acceleration outpaced the adoption and governance investment required to translate new capabilities into employee outcomes. The lesson is straightforward: investment in AI tools without commensurate investment in AI adoption infrastructure does not produce productivity gains — it produces more friction, more shadow AI, more employee rejection of the tools, and less return on the technology investment itself.
What Organisations Need to Do Differently
The WalkMe report's implicit prescription — and Kirkpatrick's explicit framing — points to three interlocking investments that organisations must make alongside, not after, AI tool deployment: building employee trust through transparent governance and demonstrated reliability; closing the communication gap on AI policy so that the 34% of workers who do not know which tools are approved can actually comply; and equipping AI tools with the contextual intelligence — real-time context, cross-application reach, and robust guardrails — that makes sanctioned tools more useful than the unsanctioned alternatives employees are currently reaching for.
At the core of the adoption failure is a measurement problem that most organisations have not yet solved: executives are measuring AI investment performance through proxy indicators (licence uptake, executive confidence, perceived productivity improvement) that do not reflect what employees experience. Until organisations close the measurement gap — building feedback loops from employee-level technology experience back into investment and governance decisions — the chasm between executive-reported and employee-experienced AI performance will continue to widen, and the 51-day friction figure will continue to rise. Explore the latest HRTech Articles for the latest tech trends in human resources technology.
Key Takeaways
- WalkMe's fifth annual State of Digital Adoption report (3,750 executives and employees, 14 countries, 1,000+ employee enterprises) finds workers lose 51 workdays per year to technology friction — up 42% from 2025 and at a three-year high — despite a 38% increase in digital investment, with 40% of that spend underperforming.
- 54% of workers bypassed AI tools and completed tasks manually in the past 30 days; a further 33% have not used AI at work at all — the data describes not adoption friction but active rejection of employer-deployed tools.
- Three chasm-scale perception gaps: AI trust (9% workers vs 61% executives — 52 points); tool adequacy (21% workers vs 88% executives — 67 points); productivity improvement (workers losing 7.9 hours/week vs 81% of executives claiming significant AI productivity gains).
- 45% of workers used unsanctioned AI in the past 30 days, 36% with confidential data — yet only 21% have been warned about AI policies and 34% don't know which tools are approved, making compliance-based governance structurally impossible in its current form.
- The friction trend reversal (from 36 days in 2025 to 51 days in 2026) proves that AI deployment acceleration without commensurate adoption infrastructure investment creates more friction, not less — and that closing the trust, governance, and tooling gaps is the prerequisite for realising AI investment returns at the employee level.