Maximizing Your Return on AI with some strategic Thoughts

The allure of artificial intelligence (AI) seems inescapable in our current business and personal life, although the hype seems to have peaked. One reason of that might be, that AI has been seen as the solution for everything and it turned out, that it cannot solve all easys looking tasks. So slipping into the "AI for AI's sake" mindset might possibly derail companies in the wrong direction.

A famous quote out of the movie "Jurassic Park" is so timeless and yet fitting to so many circumstances:

"Your scientists were so preoccupied with whether or not they could, they didn't stop to think if they should." (Jeff Goldblum starring as Dr. Ian Malcolm)

The real value of AI lies not just in implementation but, smartly, in applying strategically where every AI decision respects its potentiality of driving tangible business outcomes. The real value of AI in business goes beyond merely setting it up, it is more the strategic use of it. Thinking of the principles of delegation in Leadership/Management-Roles can be helpful. Firstly one would estimate the value of the task, the practicability of passing it on, and afterwards, put measures in place through the selection to ensure that quality work is guaranteed.

The classic "four cluster" matrix is a common tool to illustrate cohessions, so we use another one for this cause as well:

We start by drawing the two axes, one for the complexity of the task and one for the amount of creativity or judgement to be necessary. This resulting four clusters give us a rough overview, over the basic thought about how and which type of AI to use.

But this is not enough to be efficient ("do it right") in this case, we also need to be effective ("do the right thing"). So we need to think about the generated value and the costs of implementation as well, to come to an conclusion, weather to use AI or not.

The fields in the matrix are not all possible combinations, but they should illustrate the decission process.

Let`s go for some examples, to make it more tangible.

Quadrant 1: Tasks with low complexity requiring minimal creativity or judgment (bottom left, yellow)

• AI Use: Classic AI can fully automate these tasks, as they often involve routine, rule-based processes.
• Examples: Restocking notifications, sales transaction processing, basic customer query resolutions and first-level ticket processing.
• Feasibility Consideration: Typically high business value due to increased efficiency with relatively low cost and complexity of implementation.

Quadrant 2: Tasks with high complexity requiring low creativity or judgment (bottom right, red)

• AI Use: Generative AI can be used to automate tasks that involve creating new data patterns, predictions, or complex problem-solving.
• Examples: Dynamic pricing strategies, inventory assortment planning, or advanced customer segmentation.
• Feasibility Consideration: Likely to offer significant business value, but the cost and complexity could be substantial; careful planning and ROI analysis are critical.

Quadrant 3: Tasks with low complexity requiring high creativity or judgment (top left, blue)

• AI Use: Generative AI works alongside humans, mainly in creative processes or when producing novel content.
• Examples: Generating marketing copy, designing promotional materials, or creating personalized shopping experiences.
• Feasibility Consideration: Can be high-value with varying costs; assessing the ROI is essential, as generative models may require significant initial investment.

Quadrant 4: Tasks with high complexity requiring high creativity or judgment (top right, green)

• AI Use: Supports human decision-making where AI can provide data-driven insights or creative inputs, but the final judgment rests with humans.
• Examples: Product development, market trend analysis, customer experience personalization or strategic projects like the procurement of an Enterprise-Ressource-Planning-System (ERP).
• Feasibility Consideration: Potential for high business value with moderate to high costs. Human oversight is necessary to interpret AI suggestions within the broader business context.

If one decides to use or delegate tasks to the AI, there are further relevant thoughts to be made. To be more precise, it is necessary to think of the type of AI. When talking about AI, many people mean generative AI with it. Nevertheless, it makes sense to distinguish in "classic" and "generative" AI to choose the right tool for the task. The following analogy should make the difference more tangible:

Classic AI vs. Generative AI: Understanding the Difference Through the Art of Cooking

Imagine you're in a kitchen. Classic AI is like your tried-and-true cookbook—it has specific recipes (algorithms) for dishes (tasks) that have been perfected over time. It follows these recipes to the letter, every time, ensuring consistent and predictable results. You'd turn to your cookbook for dishes that require precision and tradition, like baking a classic French baguette. In the business world, this translates to tasks like data entry, scheduled reporting, or simple customer service interactions, where reliability and efficiency are key.

Generative AI, on the other hand, is akin to an innovative chef. Armed with an understanding of flavors and techniques, this chef can create new recipes, adapt on the fly, and experiment with ingredients to concoct something completely original and often, unexpected. This is perfect when you want a customized menu for a special event or to experiment with fusion cuisine. In the retail sector, generative AI is utilized for tasks that benefit from a touch of creativity and innovation—like designing personalized shopping experiences, developing new product ideas, or creating unique marketing content.

• Use Classic AI when: You need consistency, speed, and accuracy.
• Use Generative AI when: You require innovation, personalization and adaptability

By choosing the right 'chef' for your task—be it the meticulous cookbook or the creative culinary artist—you'll ensure that your AI investment is not only savory but also delivers the most delightful and profitable experience for your business palate.

Conclusion

Investing in AI is not just a technological upgrade, it's a strategic decision that should be scrutinized and justified, just like any other business investment. By evaluating the potential outcomes one can make informed decisions that align AI implementation with business value, ensuring that every ressource spent on AI works back toward their ultimate goals: growth, customer satisfaction, and competitive advantage.

Implementing AI is a journey that goes beyond the tech—it's about fostering a culture of strategic innovation. As we move forward, let's not ask whether AI can be implemented, but how it can be implemented in a way that adds true value.