Why digital transformation fails in year two (and how to fix it)

This article is part of our Opinions section, where we invite industry professionals to share their views on the most pressing technology questions of our time.

When organisations talk about digital transformation, the conversation typically centres on implementation. Projects are evaluated based on whether the new platform was deployed on time, whether users completed training, whether legacy systems were successfully retired, and whether the initiative stayed within budget. Those are important milestones, but they’re ultimately measures of project execution, not organisational transformation.

Now that many of the ambitious digital initiatives launched in 2024 and 2025 have matured, organisations are entering the evaluation phase to determine whether those investments have actually changed how work gets done. In many cases, the answer is less encouraging than leadership expected. The software is live, employees are logging in, and the dashboards look healthy. Yet beneath the surface, manual workarounds have quietly reappeared, inconsistent processes have returned, and the efficiency gains promised during implementation have begun to erode.

This gap between implementation and long-term value is becoming increasingly evident across industries. According to Deloitte, nearly 70% of organisations believe digital transformation is the single most important investment they can make to drive enterprise value. Yet many struggle to define the key performance indicators that demonstrate whether those investments are actually delivering measurable business outcomes. Gartner’s 2024 Global CIO Survey reinforces this challenge, finding that only 48% of digital initiatives meet or exceed their intended business outcome targets.

These findings point to a critical misconception about transformation. Technology doesn’t fundamentally change an organisation simply because it’s deployed. Sustainable transformation depends on changing the way people work every day, which is much more difficult to achieve.

Transformation succeeds through workflows, not software

One of the most common assumptions in enterprise technology is that introducing a new platform naturally leads employees to adopt new ways of working. 

Experience suggests otherwise. Employees don’t think in terms of applications, architectures, or technology roadmaps. Their focus is on completing their responsibilities efficiently, accurately, and with as little friction as possible.

If generating a document requires navigating multiple screens, employees will look for shortcuts. If approvals slow down, conversations revert to email. If locating the latest policy or template becomes difficult, employees rely on the colleague who “knows where everything is.” None of these behaviours is acts of resistance. They’re practical adaptations to processes that, because they are new, no longer feel intuitive.

This is where many transformation initiatives begin to lose momentum. Organisations often devote tremendous effort to selecting the right technology, configuring the platform, and managing implementation. Comparatively, little attention is given to whether the redesigned workflow is actually easier than the one it replaced.

The reality is that employees adopt workflows, not technology. If the new process reduces friction, improves consistency, and makes everyday work easier, adoption follows naturally. If it doesn’t, people create alternatives, and those alternatives gradually undermine the very efficiencies the transformation was intended to deliver.

The second year reveals whether the transformation actually took hold

Most organisations conduct formal reviews immediately after an implementation. They evaluate adoption metrics, monitor system stability, and confirm that objectives have been met. Far fewer revisit those initiatives a year or two later to assess whether the organisation is still operating as intended.

In my view, that second-year assessment is often the most valuable.

By that point, employees have settled into habits. Initial excitement has faded, project teams have moved on, and day-to-day operations reveal whether the technology genuinely supports the business or whether people have quietly found ways around it.

Several warning signs consistently emerge when transformation begins to lose traction: 

• Manual workarounds start becoming the default rather than the exception.
• Different departments solve the same problems in different ways, introducing unnecessary variation in customer experiences and internal operations. 
• Employees become increasingly unsure which documents, templates, or procedures represent the current source of truth. 

Eventually, critical processes depend less on documented workflows and more on the institutional knowledge of a small number of experienced employees.

Of these indicators, institutional knowledge concerns me the most.

Every organisation benefits from experienced employees who understand the nuances of the business. The problem arises when critical operations depend exclusively on those individuals because essential processes were never fully documented, standardised, or embedded within the organisation’s systems. That creates operational risk that extends well beyond efficiency. It complicates onboarding, slows organisational growth, increases compliance exposure, and leaves the business vulnerable whenever key personnel change roles or leave altogether.

A successful transformation should reduce dependence on institutional knowledge by making operational knowledge accessible, governed, and repeatable across the organisation.

Document operations are often where transformation quietly breaks down

While organisations invest heavily in modernising enterprise applications, they frequently underestimate the role document automation (or operations)  plays in sustaining those investments over time. 

Every business process is ultimately supported by information. Documents define agreements, policies guide decisions, templates standardise customer interactions, and approvals establish governance. When those assets become fragmented across shared drives, collaboration platforms, email attachments, or personal folders, consistency deteriorates.

The challenge isn’t simply that employees struggle to locate information. It’s when they begin questioning whether the information they’ve found can be trusted.

Once confidence in documented processes erodes, employees naturally develop their own methods for completing work. Different teams create their own templates. Business units maintain separate repositories. Procedures evolve independently, creating variations that may seem minor in isolation but collectively introduce significant operational complexity.

This is particularly problematic in highly regulated industries, where consistency is fundamental to compliance. If employees cannot reliably access the latest approved documentation or execute standardised processes, organisations increase their exposure to audit findings, regulatory issues, and inconsistent customer experiences.

Document operations are often treated as administrative concerns rather than strategic functions. In reality, they’re fundamental to operational scalability. Organisations cannot expect consistent execution if the information supporting that execution is fragmented, outdated, or disconnected from everyday workflows.

Continuous improvement matters more than continuous implementation

Organisations that sustain the value of digital transformation recognise that implementation marks the beginning of operational improvement, not its conclusion.

Business processes naturally evolve as organisations grow, regulations change, customer expectations shift, and new technologies emerge. Workflows that functioned well during implementation may introduce unnecessary complexity two years later if they are never revisited.

That is why continuous operational evaluation is so important. Leaders should regularly examine where employees experience friction, which manual processes persist, and whether approvals, documentation, and workflows still align with how the business operates today.

Equally important is measuring the right outcomes.

Login rates, training completion percentages, and system utilisation provide useful implementation metrics, but they reveal very little about operational effectiveness. More meaningful indicators include cycle times, process consistency across departments, reductions in manual touchpoints, documentation accuracy, onboarding efficiency, and the organisation’s ability to execute critical workflows without relying on institutional knowledge.

These measures provide a far clearer picture of whether transformation has become embedded in everyday operations or whether employees are gradually returning to familiar habits.

Digital transformation is ultimately a people strategy

As organisations accelerate investments in artificial intelligence, automation, and intelligent agents, this lesson becomes even more relevant. AI can automate tasks and improve decision-making, but it cannot compensate for inconsistent processes, fragmented documentation, or poorly governed workflows. In many cases, it simply amplifies the strengths (or weaknesses) already present in an organisation’s operational foundation.

The organizations realizing the greatest return from digital transformation aren’t necessarily those implementing the newest technologies first. They’re the ones continuously refining the workflows surrounding those technologies. They simplify processes rather than adding complexity. They treat documentation as a living operational asset rather than a static repository. Most importantly, they make it easier for employees to work within approved systems than to work around them.

Digital transformation has never been solely about modernising technology. It’s about creating operational systems that enable people to perform their work more consistently, more efficiently, and with greater confidence than before.

Technology may initiate transformation, but people determine whether it lasts. And that’s why the most important measure of success isn’t what happens at go-live. It’s what your employees are still doing two years later.

Related articles

• Matt Scavetta, CTIO at FutureTech: “Agentic AI is different. It forces organizations to think more like software companies.”
• Process mining’s next job is to stop AI agents automating broken work
• Ananda Iyer, Head of Customer Success at Annexa: “The technology has to be built for change”
• Why change management is the real driver of digital transformation