Insight first, then acceleration: the key to a successful AI organization

As an organization, you can gain a lot of benefits from AI. But only if you realize that AI is a means to an end, and not an end in itself.

Many organizations do not see this. As a result, AI initiatives lead to disappointing results, because the expected value is not forthcoming. Research  shows that only 11 percent of organizations achieve a clear return on AI investment (ROI).

There is a laundry list of reasons for this. Think of ethical issues, a lack of trust or insufficient AI literacy within the organization.

Understandable, but also a shame. Because you can indeed run a successful AI organization. How? By reasoning from value. The fundamental question is: 'How do we achieve our organizational goals with the use of AI?' If that is your starting point, you have a big advantage over organizations that are rushing to roll out AI initiatives "because you have to do 'something' with AI".

The usefulness of a baseline measurement

If you want to accelerate with AI, it is important to first have insight into where you are and where you want to go. You gain that insight by performing a baseline measurement. You will take a close look at all kinds of facets and answer essential questions.

For example, do you have a clear overview of your business objectives? Do you have sponsorship from the management? Are employees ready for AI?

In addition, it is crucial to think about rules around AI use. When these are lacking, employees often dare to experiment less. After all, they do not know what exactly is allowed. For example, if you haven't clearly defined whether people can use personal data in Copilot, they'll likely be inclined not to. But if your organization already has this data in Word, it doesn't really matter. Then the information is already in your Microsoft 365 environment.

Another question you would like to ask during a baseline measurement is: 'Are employees already using AI?' Perhaps some are already building AI agents to work more efficiently and do not understand that they have already laid an 'AI agent foundation' for the organization (whether desired or not)!

Of course, as an organization, you also take a critical look at data security. Because the last thing you want is for people to start working with AI and encounter a serious security problem. However, it is important to realize that it is impossible to tick all the boxes before you start AI initiatives. Then you are two years further and there are new requirements that you have to meet. You can never start like this!

A better idea is to make Copilot available to a small group of employees first, for example. You prepare these pioneers for the possibility that Copilot can uncover a problem around data security. Let them report this immediately, so that you can take action to improve your security. This is how you use Copilot to strengthen security!

Getting started constructively with AI

In short, a baseline measurement is indispensable. But it will only be useful if you also map out a few other things: what your ambition is, what you need in the field of AI and where the gap is that you have to bridge to achieve your objectives.

Only then can you determine which steps are needed to realize your ambition. And so you can work constructively with AI to get real value out of it.