Tech Reality Check 2026: AI optimisation is just the beginning

The Tech Reality Check 2026 is a Conclusion study among IT and business decision-makers in the Netherlands, Germany, Spain, and Portugal. Its main message is sharp and largely justified: according to the report, the biggest gap lies in decision-making. Technology is widely available; control and direction lag behind. Organisations invest, recognise risks, and launch initiatives, but still translate too little of that into effective management.

We agree with most of that. The report also notes that AI is currently used mainly to make existing processes more efficient, while fundamentally different ways of working and new products or services are less visible. We would like to add one point: optimisation is often exactly where transformation starts.

So the question is not whether optimisation is enough, but what you do with the room it creates.

Optimisation is the foundation, not a partial solution

Much of what we implement for clients does indeed involve repairing or accelerating existing processes. An invoice flow that has been manually retyped for years. A knowledge process where nobody can find what was decided last year. A request that passes through four departments before anyone acts on it. AI removes the friction.

Optimisation alone can indeed feel like settling for efficiency and nothing more. In our work, we see a great deal of value created precisely there. A broken process that finally works delivers something immediately: for the team that hates the work, for the client who gets an answer sooner, and for the margin.

The risk only arises when optimisation becomes the end point. A repaired process that leads nowhere afterwards remains mainly a more efficient version of the same work.

The difference lies in how you use that gain. Treat a cleaned-up process as a starting point: the time and capacity freed up are exactly the room you need to take the next step. An organisation whose processes are not in order never gets to that step. It only builds a bigger mess faster.

Much of the work right now lies in optimisation and repair. That is not a sign of weakness. It is where the real value is today, and the foundation for everything that follows.

The real opportunity: accelerating what was previously impossible

That “afterwards” is the most interesting part. Alongside repairs, AI makes something new possible: products, processes, or activities that previously simply could not happen because they were too expensive, too slow, or too labour-intensive.

An advisory team that normally handles three files a week, and will soon prepare thirty. A service you could previously offer only to large clients, which suddenly becomes profitable for SMEs too. An analysis you previously skipped for lack of time, which now runs as standard. Then you are selling something different from before.

That is where transformation becomes concrete: not because someone at the top suddenly invents a new business model, but because organisations discover more and more places where they can remove the brakes. What you can offer expands there, not just how quickly you deliver it.

The most useful starting point is a simple question to ask yourself: which task would you do ten times more often if it cost almost nothing? That is usually exactly where acceleration creates something new rather than just saving time.

Why decision-making makes the difference

Whether you are repairing a process or introducing a new service, the decisive difference rarely lies in the technology you choose. Gartner expects over 40% of agentic AI projects to be cancelled by the end of 2027, because of rising costs, unclear business value, or inadequate risk controls. The technology itself is rarely the cause.

We recognise that. Most AI projects that get stuck do so because of the surrounding process and the choices nobody makes. Who owns it? Which decision are we actually automating? Who intervenes when things go wrong? That is exactly the gap the report identifies.

We would add one thing, because we keep seeing it in our work: a decision-making gap is fundamentally a change management issue. Change management has always mattered in every software or IT project. With AI it carries more weight, because AI cuts across departments and affects how people do their work.

Prosci, known for the ADKAR change management model, has researched this since 1998. The figures are clear: projects with active, visible executive sponsorship achieve their goals in 79% of cases, compared with 27% where sponsorship is weak. Whether senior leaders visibly support the change is consistently one of the most important success factors. It is the same decision-making gap, one layer deeper.

What this means in practice

The report’s realism is justified. Repair the broken processes, because that generates money now and provides the foundation. Then use the room freed up to find where AI makes something possible that was previously too expensive, too slow, or too labour-intensive. Both require the same discipline: clear decision-making and leadership that truly supports the change.

Want to know where those two opportunities, repairing and accelerating, lie in your own processes? Our AI Readiness Scan is the starting point. In ten minutes, you see which processes are ready to be cleaned up and where the larger opportunities lie.

A note on how this was made: our content is AI-assisted. We use AI for research and drafting a first concept. A human reviews and edits the text and takes responsibility for everything we publish. Read how we make our content.