"Don't waste time on AI for AI's sake. Be motivated by what it will do for you, not by how sci-fi it sounds."
Despite recognizing the importance of integrating AI capabilities to stay competitive, businesses do not know how to jumpstart their AI journey. Some companies have already done prototypes but now struggle to define and prioritize AI use cases. Good for them! The real problem arises for the many enterprises that grossly underestimate the challenge, feeling they have done the same thing many times before. After all, everyone went through Digital transformation programs. Also, many organizations now know how to leverage Big Data, and its use cases and a common assumption is that AI is no different.
The results are failed prototypes; some of them even reach a good demo, followed by disappointment. After initial success at the demo, Product owners fail to justify the investment and value creation, practical implementation hurdles and fail to scale. Often, there is a big gap between the promise made to the customer without considering all aspects, limitations and what is delivered. In all these cases, high hopes of “value creation” and “value capture” rarely emerge.
For all the reasons mentioned above, we need a framework that can help us decide if there is a need to use machine learning in your product. Is your use-case suitable for the application of AI?
We can think of this framework as an AI value creation framework at a high level. There are potentially three areas that need to be considered. Your use case should qualify at least one of the below to be considered a good candidate for machine learning.
- Increase Efficiency and Reduce cost through automation: AI application should reduce costs by achieving operation excellence by leveraging automation for tasks that could not be automated otherwise and involves data such as automated image tagging. Automation makes the process more efficient and cost-effective. For example, back-office automation of a banking institution leveraging a human-in-loop credit card approval system.
- Customer Engagement: AI applications should help in either increasing number of customers (low-cost and time-efficient operations; serving more customers and increasing market share) or increasing customer intimacy (Customer engagement experience). AI applications are more efficient at performing routine tasks involving a fixed set of functions, with significantly fewer errors than humans. It saves time and makes customers happy about your product offerings, resulting in a premium service providing more revenue. For example, real-time voice pattern tracking of a customer care call and offering feedback to the call centre executive make the conversation more productive, resulting in an empathetic and happy customer and engaged worker, resulting in more business.
- New business value: AI can also help businesses find new opportunities by combining complementary digital technologies such as cloud and mobile. This new business value can be captured by either reaching the new customer segments that were unexplored so far or creating an entirely new product, or building a new business model. For example, Manufacturing companies can leverage AI and data to perform predictive maintenance on sophisticated machines.
This framework should provide guiding principles for the product owners and leaders while evaluating new opportunities. Not all use cases need the application of machine learning, and not all machine learning applications can be scaled equally. Each problem is unique and should be considered case-by-case by evaluating what value it creates for the customer and what value you as an organization can capture by providing these services. This framework can help you to evaluate and prioritize opportunities.
Other Suggested readings:
- Winning with AI; MIT Press review
Share this Post