AI transformation is not primarily a technology problem

For many companies, the conversation around AI still starts with tools, infrastructure and implementation. Which model should we use? Which platform should we buy? How should we integrate AI into existing systems?

According to Maxwell Minckler from Google’s EMEA Market Insights team, those questions matter, but they are not the whole story.

In a recent Ecommerce Suomi Podcast episode, Maxwell shared findings from Google’s research into AI transformation across more than 2,000 companies worldwide. The research looked at why some companies are moving ahead with AI while others are still struggling to turn experimentation into everyday business value.

One of the clearest findings was that the difference is not explained by technology alone.

The companies making the most progress tend to have different habits, ways of working and attitudes towards AI.

Most companies are still figuring it out

AI transformation is still at an early stage for most organizations.

Maxwell explained that only a small share of companies have reached a point where AI is used as part of normal day-to-day business. A larger group is moving towards that stage, but the majority are still somewhere earlier in the journey.

That matters for companies that feel they are already behind.

The current situation is not one where everyone else has completed their AI transformation. Most companies are still working out how AI should fit into their business.

At the same time, the companies that have progressed further are reporting stronger business performance. That makes the next question especially important.

What are they doing differently?

Three common mindsets around AI

The research identified three recurring mindsets in how people and teams talk about AI.

The first is FOMO, Fear of Missing Out.

This is the feeling that competitors are already moving and that your company needs to catch up quickly. FOMO can be useful because it creates urgency and gets organizations started.

It can also lead to poor decisions.

A company driven too heavily by FOMO may introduce AI without a clear business objective, copy what competitors are doing or focus on proving that employees are using AI rather than asking whether that usage is creating real value.

The second mindset is FOMU, Fear of Messing Up.

If FOMO pushes companies forward, FOMU can hold them back.

Concerns about risk, mistakes and poor decisions are necessary, particularly when AI is being introduced into important business processes. Problems arise when risk avoidance becomes the main driver.

Companies may then limit AI to small pilots at the edge of the organization instead of looking at the parts of the business where it could make the biggest difference.

The third mindset is FOMA, Focus on Maximizing Advantages.

This was more common among the companies leading in AI transformation.

The important point is that these companies are not free from fear. Their employees still experience FOMO and FOMU.

The difference is that their teams are better at moving the discussion towards a more useful question:

Where can AI create the most value for our business and our people?

Leadership language matters

How leaders talk about AI can influence how employees respond to it.

One example Maxwell raised was the familiar phrase “human in the loop.”

The idea is sensible. AI systems should not operate without appropriate human oversight.

But the wording can also unintentionally reduce the role of the employee to someone who simply checks and approves what AI has produced.

Maxwell suggested a more useful way of thinking about it:

Own the loop.

The employee should not just remain somewhere inside the process. They should retain ownership of the process.

This changes the role of AI from something that replaces responsibility into something that supports the person responsible for the work.

Explain why AI is being introduced

Another important leadership principle is what Maxwell calls “frame the why.”

There can be a gap between what companies want from AI and what employees want from it.

Organizations often focus on efficiency. They want processes to become faster and cheaper.

Employees may be looking for something different. They want to improve the quality of their work, develop their skills and become better at what they do.

If leadership only talks about efficiency and cost reduction, employees can easily interpret AI as a threat.

The message becomes:

“How can we do the same work with fewer people?”

A more effective approach is to explain how AI can improve both the business and the work itself.

AI can reveal where human value really sits

Maxwell used his own profession, market research, as an example.

AI can automate many operational parts of research. It can support survey work, quality assurance, briefs and other repetitive tasks.

If a researcher defines their professional value through those tasks, automation can feel threatening.

But removing some of that operational work can also make something else more visible.

It can show where the real value of the role has always been.

For a researcher, that may be understanding customers, finding useful insights, supporting better business decisions and building relationships with stakeholders.

In that sense, AI can act as a revealer.

It can help separate the routine parts of a role from the parts where human expertise creates the most value.

From individual tools to agentic workflows

For ecommerce companies, one of the most important parts of the discussion was what comes next.

The first phase of AI adoption has largely focused on individual tools.

Employees use AI to write emails, summarize documents, conduct research or speed up individual tasks.

That is a useful starting point.

The next phase is broader.

Companies are beginning to look at entire workflows and ask where AI agents could take responsibility for parts of those processes.

This requires companies to understand their own work much more clearly.

Processes need to be documented. Information needs to be structured. The steps inside a workflow need to be understandable not only to employees but also to machines.

Maxwell described this as making work more machine-legible.

Companies that understand their workflows and can describe them clearly will be in a stronger position to build agentic processes around them.

What should ecommerce companies focus on now?

The practical takeaway is not that every ecommerce company needs to adopt the newest AI tool immediately.

A better starting point is to look at how work is currently done.

Which processes are repeated every day?

Which parts require real human judgment?

Which parts are mainly operational?

Where is information stored?

Are the important workflows documented clearly enough that an AI system could understand them?

These questions are becoming increasingly relevant as AI moves from individual productivity tools towards more autonomous workflows.

Technology will continue to develop quickly.

The companies that benefit most may not simply be the ones that buy the newest tools first. They may be the ones that understand their own business well enough to know where AI actually belongs.

Listen to the full conversation

Listen to the full Ecommerce Suomi Podcast conversation with Maxwell Minckler from Google:

YouTube:
https://www.youtube.com/watch?v=kX3yISrPam4

Spotify:
https://open.spotify.com/episode/24BFvX4dNwCvWLTyq6qWFl?si=JactGW5BT2G2O3mTM_kIrw