Does AI implementation only begin after buying AI?

One question kept coming back to me as I reviewed the latest research on AI implementation. Why do organisations that invest in very similar AI technologies often achieve completely different results? The usual explanations are familiar: data quality, governance, leadership, change management and training. They all matter, and the evidence supporting them is convincing. Yet it felt as though something was still missing.

Perhaps we are looking at the wrong starting point.

Most organisations assume AI implementation begins when they approve the investment, choose a supplier and start deploying the technology. Everything before that is treated as normal business. But by the time an AI project officially starts, the organisation already has its people in place. Those people have been recruited, onboarded and developed within a particular organisational environment. They have learned how decisions are made, how change is managed and which behaviours are encouraged. By the time the first AI platform arrives, the workforce has already been shaped.

That simple observation led me to ask a different question. What if organisations do not begin AI implementation with the same level of organisational capability? Two companies may buy the same technology, invest similar budgets and follow comparable implementation plans, yet achieve very different outcomes because they were never equally prepared in the first place.

That question became the foundation of my latest SSRN working paper. Instead of focusing only on technology readiness, I explored whether organisations also need workforce readiness before implementation starts. This led to the concept of Pre-Implementation Organisational Readiness (PIOR). The idea is straightforward. Technology readiness asks, "Can we deploy AI successfully?" PIOR asks, "Have we already prepared our people to work successfully with AI?"

The distinction may seem small, but it changes the conversation completely. Recruitment is no longer simply about hiring someone who can perform today's job. It becomes an opportunity to bring into the organisation people who are willing to learn, adapt and work alongside technologies that will continue to evolve. Likewise, onboarding becomes much more than explaining company policies. It is where organisations begin shaping how employees think about AI, professional judgement, responsibility and continuous learning.

The paper illustrates this by examining a manufacturing organisation that introduces generative AI across several business functions. The technology works well. Pilot projects deliver encouraging results, and leadership approves a wider rollout. Yet, a few months later, adoption looks very different across departments. Some teams quickly integrate AI into everyday work, while others rarely use it. When the organisation investigates, the technology is not the problem. The real differences already existed before implementation began through recruitment decisions, onboarding practices and management expectations.

This does not mean technology is less important. On the contrary, technology readiness remains essential. The paper argues, however, that it may only represent part of the picture. Sustainable AI adoption appears to depend not only on the technology itself but also on the organisational capability that has been quietly developing long before AI enters the business.

Perhaps AI implementation begins much earlier than we think.

Perhaps it begins with the people an organisation hires, how it integrates them into the organisation and how it prepares them to learn, adapt and work confidently alongside AI. If that assumption is correct, organisations may need to rethink not only how they implement AI, but also when implementation truly begins.

Want to explore the idea further?

In the full SSRN working paper, you will find:

• Why AI implementation may begin before technology is purchased.

• The concept of Pre-Implementation Organisational Readiness (PIOR).

• How workforce readiness complements technology readiness.

• An executive example illustrating the framework in practice.

• Future opportunities for research and organisational applications.

If the topic interests you, the SSRN link is below. I hope the paper offers a different way of thinking about AI implementation and proves useful for anyone involved in AI strategy, organisational transformation or preparing people to work successfully with AI.

SSRN pdf