Digital transformation is the heart of discussions in almost every field these days. Organizations the world over are taking an active interest in AI-driven tech, viewing it as the key to productivity gains, and understandably so, given the depth of its potential. However, despite the ambition to implement it, when it comes to scaling up the implementation of AI, progress has been relatively slow in places.

In its recently published report, The State of AI in 2025, McKinsey found that of those organizations surveyed, 88% are actively using AI on a regular basis in at least one business function – up 10% from the previous year. While that is a clear sign of positive intent, close to two-thirds of the respondents in that same survey reported that their organizations have yet to scale up AI. It's something of a reality check – while companies are clearly more than willing to embrace the latest tech, many are still in the experimentation stage.

So, just what's at the heart of this, and how can enterprises avoid becoming trapped in pilot phase purgatory?

Productivity gaps are the elephant in the room

In my view, one key factor at play is the discrepancy between the perception of AI and the realities of implementing it. With the huge developmental leaps that have been made in recent years, a lot of excitement has been built around AI tech, such that it is often presented as a magic bullet – a fix-all that will transform a company overnight. But the truth is that it's not a plug-and-play solution, and the introduction of AI does not automatically mean greater productivity or efficiency.

The rise of AI tech has been so meteoric that there has been relatively little time to become familiar with the intricacies of its utilization. At the same time, the competitiveness of modern markets has put companies under pressure to embrace it, often before they're ready to do so. The findings from McKinsey's research indicated as much.

Moreover, there is a tangible productivity gap emerging where AI is being implemented. A recent survey from EY found that while 64% of respondents reported increases in their workloads in the last year, only 5% are actually using AI to meaningfully impact their workflows. EY concluded that companies are missing out on as much as 40% productivity gains with AI because employees simply aren't ready for it. Having seen the challenges firsthand, it's hard to disagree.

When employees lack the competencies to utilize AI tech comfortably, things start to get messy. Unfamiliar interfaces and features slow down productivity, and if staff can't remedy that issue quickly, they begin to seek out workarounds. This makes workflows convoluted and inefficient, and can result in the proliferation of shadow AI. Aside from introducing additional security concerns, this drastically reduces the ROI of new technologies. Stakeholders then become sceptical and may even cut their losses altogether, and before you know it, the company is back to square one.

It's not the technology that's failing – it's the approach to implementation.

Digital adoption is the key

AI can be an engine for growth, but as with any engine, it is just one component of a machine, and it can only function in the right context and under suitable conditions. To be genuinely impactful, AI tools need to be assimilated into an organization that's set up to utilize them, and that's where digital adoption is critical.

Amid all the buzz of digital transformation and AI, a lot of companies have forgotten to consider what facilitates the transition to new technologies. This is what digital adoption focuses on, and why it's so important. It's all about the strategic introduction of new systems, and establishing conditions for successful implementation by creating a company culture that's actually ready for those systems.

In devising digital adoption strategies, those seeking to implement AI tech must prioritize onboarding, first and foremost, to address the skill gaps at the root of productivity issues.

Making onboarding adaptive and intuitive

It is, admittedly, easier said than done to upskill enough to utilize the kind of AI tech we're talking about. The complexity of modern enterprise tools can create something of a steep learning curve, but there are tools designed to simplify things.

Digital transformation specialists like WalkMe, for instance, have developed purpose-built digital adoption platforms (DAPs), which provide contextual guidance and prompts overlaid on the UI of other tools. This can make the onboarding experience feel more personalized, more intuitive, and less stressful for employees, which reduces the risk of abandonment. Moreover, DAPs typically have built-in analytics and reporting capabilities, so companies can track the usage of new tools and find ways to alleviate friction.

In conjunction with onboarding tools, companies need to establish open communication channels for employees to provide feedback on what they're being asked to learn. What's more, adoption needs to be modelled from the top down to ensure cohesive messaging that encourages unilateral buy-in. This is all part of creating an internal culture that is ready to truly assimilate the new tech.

Digital transformation isn't easy, but nothing worth doing ever is. With a smart, employee-centric digital adoption strategy supported by adaptive onboarding tools, it's absolutely possible to bridge the skill gaps that are preventing companies from scaling their AI implementation.

Final thoughts

AI has quickly become a focal point of almost all technical ambitions these days, but simply introducing it into our organizations is no guarantee of success. To get true value from the tech, we first need to invest in the people who will engage with it day in, day out. By fostering readiness with digital adoption strategies that streamline onboarding and promote agile company culture, it's possible to unlock the scalable potential of AI.