The difference between average and outstanding AI

Sometimes less is more, and simplicity explains otherwise difficult subjects.

The way some companies scale value from their investments in AI is one such topic.

Organizations other than those featured in profiles like this remain challenged to see past the superficial features of those reportedly winning with AI: that they are very large companies with data science teams and money to burn on experiments and productionized use cases.

Therefore, today I thought I would attempt a simple explanation of what these companies have done, that almost any organization can learn from and begin to improve.

If you were to dig into the examples, research and countless opinions, you would find that success has nothing to do with data science or your bank account.

Certain companies experience a value compounding effect with each successive use case they execute. Others do not and in some cases walk away from AI entirely. One way to visualize the difference is like this:

The chart on the left represents how most companies experience the benefits of AI today; one project or use case at a time, apart one another in different areas of the business. Sometimes a high profile project with transformational intentions fails to meet expectations, and the executive team concludes that AI is a failed experiment.

The company on the right took a different approach. Somehow beyond the initial use case, additional projects delivered extra value. Sometimes only a bit more value, other times a lot more. After a while, the total value is off the chart. The CEO notices. AI is an element of the company’s digital transformation.

The point is to recognize the best practices embodied in that white space with the question mark. Rather than tell you what I know, think, or can cite to that end, I will leave it to anyone who cares to offer their point of view.