Digital transformation often starts with good intentions. A company wants better visibility, faster workflows, stronger compliance, improved productivity, or smarter decision-making. So it buys a new platform, launches a dashboard programme, automates approvals, connects systems, or rolls out AI tools.
But there is an uncomfortable question many organisations do not ask early enough: are we actually solving the problem, or are we just digitising it?
That question matters because technology does not automatically improve operations. Recent McKinsey and BCG work points to a recurring pattern: technology programmes struggle when organisations fail to manage interdependencies, redesign how work actually happens, or build the capabilities needed to turn digital investments into performance gains.
A bad process in a new system is still a bad process
One of the most common mistakes in digital transformation is assuming that software will fix process weakness. It will not. If a workflow is bloated, unclear, over-controlled, duplicated, or based on poor ownership, putting it into a digital system usually makes it faster to execute but not better in design.
In some cases, digitisation actually hardens the problem. What used to be a messy manual workaround becomes a formal, system-driven process that is harder to challenge because it now looks professional, traceable, and modern. The business mistakes “digitally managed” for “operationally improved.” That is exactly why large-scale tech programmes often drift: they tackle systems without resolving the deeper operational interdependencies underneath them.
Digitisation can make waste less visible, not less real
Poor processes do not disappear when they become digital. They simply change form.
Waiting becomes digital queueing. Rework becomes repeated data entry. Overprocessing becomes excessive workflow steps and approvals inside software. Lack of clarity becomes more dashboards, more notifications, and more reporting layers. Fragmented ownership becomes multiple systems with overlapping responsibilities.
The organisation may feel more advanced because work is now happening through platforms, forms, bots, and analytics. But the hidden waste is still there. Sometimes it is worse, because digital systems can scale poor design more quickly across the organisation. BCG’s 2024 work on large-scale tech programmes highlights that failure to define and track interdependencies is a major reason programmes run behind schedule and over budget.
The warning signs are usually easy to spot
Most organisations can tell when they are digitising problems instead of solving them, even if they do not use those words.
The signs are familiar. A new system goes live, but users still keep spreadsheets on the side. A dashboard exists, but meetings still revolve around arguing over the numbers. Workflow automation is in place, but cycle time has not meaningfully improved. AI-generated outputs appear impressive, but teams still do not trust the decisions enough to act on them. The system is technically live, yet operationally fragile.
That gap between deployment and value is where many transformation programmes stall. McKinsey’s 2025 and 2026 operations coverage stresses that productivity gains come from combining technology with cross-functional collaboration, capability building, and real operational rewiring, not from deployment alone.
Automation is not the same as simplification
A useful discipline is to separate three things that organisations often blur together: digitisation, automation, and improvement.
Digitisation means converting work into a digital form. Automation means reducing manual effort through system logic or machine execution. Improvement means making the process better.
These are not the same. A company can digitise a terrible process. It can automate unnecessary approvals. It can build AI on top of inconsistent data. It can add advanced tools to a system that nobody fully owns.
If simplification does not happen first, automation can accelerate waste instead of removing it. BCG’s 2025 work on smarter tech investment makes a similar point from a complexity angle: simplifying products, platforms, and processes frees resources and improves the odds that transformation creates lasting value.
The AI era raises the stakes
This issue becomes even more important with AI.
When organisations layer AI onto weak processes, they do not become intelligent by default. They can become faster at producing low-value outputs, more confident in flawed assumptions, and more dependent on data structures that were never designed well in the first place. McKinsey described 2025 as a turning point when generative and agentic AI moved from experimentation toward enterprise impact in operations, which means the consequences of weak process design now scale faster than before.
That makes process thinking more valuable, not less. Before asking “How can AI help this process?” leaders should ask “Should this process work this way at all?” and “What is the real decision, handoff, or failure point we are trying to improve?”
Industry 5.0 pushes the question further
The Industry 5.0 perspective adds another useful challenge. The European Commission frames Industry 5.0 around human-centricity, resilience, and sustainability, not just digital advancement. In that model, a transformation is not successful simply because it introduces more technology. It has to improve the wider system: how people work, how resilient operations are, and how responsibly value is created.
That matters because many digitised problems are not only inefficient. They are also frustrating for workers, brittle under disruption, and wasteful in how they consume time, energy, and attention. A process that is faster but more confusing is not necessarily better. A system that is more automated but less resilient is not necessarily more mature. Industry 5.0 reminds organisations that progress should be judged by the quality of the operating model, not the quantity of the technology.
Why organisations fall into this trap
There are a few reasons this happens repeatedly.
First, software is visible. Process redesign is harder to see and harder to sell internally. A system launch looks like progress. Rethinking decision rights, reducing approvals, clarifying ownership, or removing non-value-added steps feels slower and less dramatic.
Second, many organisations separate digital teams from operational teams. One group owns the tools, another owns the process, and a third owns change. The result is fragmentation. The technology gets implemented, but the operating model remains largely untouched. That is one reason both McKinsey and BCG keep returning to cross-functional coordination and interdependency management as major success factors in transformation.
Third, leaders sometimes confuse data abundance with process understanding. More reports, more dashboards, and more alerts can create the impression of control while masking the fact that the underlying workflow still makes little sense.
Better questions lead to better transformation
To avoid digitising problems, organisations need to ask better questions before they buy, build, or automate.
What is the actual problem?
Where does the waste really sit?
Which approvals, handoffs, or controls add value, and which just add delay?
What decisions are people struggling to make today?
What workarounds already exist?
If we removed the software from the picture, would this process still make sense?
Those questions sound basic, but they are often skipped because the organisation is eager to move into solution mode. Yet the quality of transformation depends heavily on whether leaders diagnose before they digitise.
What good looks like
Good digital transformation does not start with “what system do we need?” It starts with “what outcome do we need, and what process design will support it?”
That usually means simplifying first, clarifying ownership first, defining critical decisions first, and building capability alongside technology. It also means treating digital tools as enablers of better operations rather than substitutes for operational thinking. Recent McKinsey work on digital skill building reinforces that digital performance now depends on wider workforce capability, not just specialist tech teams.
The strongest programmes do not just digitise existing work. They challenge whether the work should exist in that form at all.
Conclusion
If a process is unclear, wasteful, fragmented, or poorly owned, digitising it will not solve the real problem. It may make the problem look cleaner, faster, and more professional, but the weakness is still there.
That is the core risk in many transformation programmes today. Companies invest in technology when what they really need is better process design, better cross-functional thinking, and better judgement about how work should flow. In a world of AI, automation, and Industry 5.0, that discipline matters even more. The future does not belong to organisations that digitise everything. It belongs to those that improve the right things before, during, and after digitising them.