Artificial intelligence and machine learning are not just for technology companies anymore. The COVID-19 pandemic saw to that as organizations around the world enacted work-from-home policies and other precautionary measures to protect employees while keeping them productive.
New coronavirus variants aside, the economy is becoming more dynamic again. Yet elements of the economy are also changing, like the adoption of AI, presenting new challenges for those at work. To understand how artificial intelligence is helping employees to transition and adapt within the changing world of work, Digital Journal heard from Jordan Reynolds, Global Director of Data Science, Kalypso.
According to Reynolds: “Economists are divided on whether current inflation rates and labor shortages will last 10 months or 10 years, yet no one can deny that the global economy is changing dramatically. Factors such as bottlenecks and hiring difficulties have contributed to the present strain on labor, resources and the supply chain as a whole, which some experts say could last longer than originally anticipated. Simultaneously, the older population is exiting the workforce, leaving gaps in knowledge and more headcount concerns in some industries.”
Digital Journal: How can companies adapt, given the turmoil of the past two years?
Jordan Reynolds: Efficiency has never been more imperative—on one hand, to meet increased global demand, and on the other, to comply with prolonged social distancing and other health-related mandates. Where does this leave the modern business decision maker? Artificial intelligence (AI), for one, is a wise investment and one that has already helped many. But it too comes with parameters to consider. To realize sustainable and successful implementation of AI long past these pandemic years, it’s critical to lay a solid foundation that aligns with real business challenges and takes a scalable approach.
DJ: What can AI truly accomplish?
Reynolds: Many businesses are apprehensive about AI, assuming it may not have a sensible application for what they do. But the truth is, AI is the logical next step for most organizations today. Accelerated by the pandemic, many companies are using AI to help employees work more efficiently from their homes, help factories operate with fewer personnel and give back time to staff through automation and digital transformation.
When AI is used to help solve real, compelling business challenges, companies can realize its value quickly. In the world of consumer goods, for example, AI can accomplish expediting speed to market, collecting consumer feedback on a certain product line or removing redundancies in a distributor relationship to cut operating expenses. Tying digital transformation to a real business challenge will help lay a solid foundation for future change.
DJ: What are the key actions and priorities for businesses?
Reynolds: The most successful implementations of AI in real business settings often begin with small snippets of the greater operation. As with most things in life, taking a walk-before-we-run approach is wise at the beginning of the AI journey. It’s smart to determine what AI can accomplish at the micro-level first and then scale those successes and learnings to other similar areas of an organization.
Every business, regardless of industry, has that one pressing business challenge it must address. It’s important for businesses to be thoughtful and intentional about what that need may be and avoid deviating from those primary goals for something big and shiny. Consider a biotech company that is approaching a major merger and acquisition; its immediate business goal is to accelerate the integration of its two entities.
In the future, the company will establish future infrastructure for growth, but its immediate attention must be directed to expediting the merger in order to minimize unwanted business expenses. Another example might be a major consumer goods retailer that was forced to reduce its staff during the pandemic but is now experiencing a spike in demand. It must find the most efficient way to fulfill new orders, so it turns to AI to optimize basic components within its manufacturing operations. Narrowing in on one specific challenge rather than trying to tackle everything at once will help organizations gain real value from AI quickly and sustainably.
DJ: When you say “Depth of Value Beats Breadth of Application”, what do you mean by this?
Reynolds: Too often, an AI implementation fails because it’s too much, too soon. It’s important to set priorities and focus on one key business challenge that AI can transform for a business. A minimum viable product (MVP) approach is based on this idea that depth of value beats breadth of application. The MVP approach isn’t focused on meeting the needs of every customer or every use case, but rather satisfying the needs of one specific group of customers or use case.
Companies that focus on an MVP will be able to build a full stack of capabilities and value focused on a single, defined, high-priority use case. It starts by building foundational connectivity and data models and including descriptive analytics, root cause analysis, predictive modeling, even prescriptive modeling and autonomous optimization. When value is realized for a specific high-priority use case—ideally linked to an important business challenge—companies can then scale for additional applications.
AI is a powerful tool in the right instances with the right approach—that is, when applied to real business challenges and kept focused on a narrow, high-priority use case prior to scaling to larger areas of a business. When these considerations are kept in mind, businesses will find that AI algorithms, data and infrastructures are able to operate well at different speeds and complexities, making it a valuable tool for many modern applications. Considering the unpredictability of the global economy, the scalability and agility that AI delivers has never been more critical.
