On March 19th, all residents in California, 40+ million people, were asked to “shelter in place” and leave their homes only for basic necessities. Any bay area citizen who has lived through often nightmarish commutes can now travel corridors with ease that a month ago would have been congested with bumper to bumper traffic. In the days since, New York City, Philadelphia, Ohio, Delaware, and numerous other states and cities have followed suit. Of course, these steps are necessary if we are to have any impact on the tragedy of thousands of lives lost and the hundreds of thousands that will be infected. Even with harsh measures the time needed for families and communities to heal from the COVID-19 pandemic will be measured at the very least in months. In this context of human suffering, worrying about how enterprises will react to a slowing economy might seem like, at best, an anecdotal discussion that has little if any value. The impact of COVID-19 on the world’s economy, however, has the potential to turn a healthcare crisis into an equally significant financial crisis. Unemployment numbers in the US are growing at double-digit rates each week. In China, the COVID-19 outbreak has already led to the first economic contraction since the 1970s, and the US DOW JONES stock index has dropped by nearly 35% in three months. The US economy is clearly headed for a slowdown, and enterprises worldwide, across multiple industries, are bracing for a recession.
AI & Machine Learning a Success Story at Risk?
Artificial Intelligence (AI) and Machine Learning (ML) projects have been near the top of the investment priority list for enterprises in the past five years. According to Gartner, 14% of organizations have already adopted AI, and 48% more are considering adoption by 2020. The reasons for the rapid adoption and continued interest in AI and ML are rooted in the huge impact investment in AI/ML technology can have on just about any business. A large multinational bank was able to double the close rate for new clients by deploying AI/ML-based predictions to identify target clients that fit a pattern more likely to close. The same organization increased close rates by also predicting when a client might be interested in purchasing services based on interests in other already purchased products. An insurance multinational was able to achieve an increase of 2.9X in upselling probability by focusing on customers with group contracts and on contracts that showed a high pattern of visits on their websites. These are just some use-cases where AI/ML has provided significant benefits for both consumers and enterprises. For most organizations, even during financially sound economic times, the challenge of AI/ML does not lie in measuring return on investment, but rather in the significant timelines involved in project developments. Given shrinking economies, likely layoffs, and recessionary times in our future, what happens to existing and possibly new AI/ML investments? How should businesses respond and adapt?
AI/ML In a Downturn, Help or Hindrance?
Economic downturns always bring a great deal of uncertainty — will the recession be short and sharp? Less severe but with a longer timespan? Planning for an economic downturn is, under the best of circumstances, a complex proposition. Many organizations react too slowly at the beginning of a downturn and then are not able to respond swiftly enough when the economy invariably recovers. Whenever discussing economic downturns, enterprises must begin to consider reductions in investments, but what is the right move? Especially with regards to investments in AI/ML? The real returns from AI/ML projects are typically not measured in weeks, but much longer time-frames. AI/ML projects like customer churn prediction, loan default monitoring, marketing campaign optimization, and others like them have, historically, focused on highly strategic areas of the organization targeting growth or risk management, making them high-value opportunities that can provide great yields, but that also require significant capital and resources. These AI/ML use-cases in the enterprise are even more valuable during economic downturns, the real question, then, is not whether investments in AI/ML should be re-evaluated, but rather, how can enterprises continue to advance their AI/ML initiatives during an economic downturn? As organizations re-assess their investments during a slowing economy, one of the most likely responses will be to slow down – or even halt – hiring of new talent – especially highly skilled, more expensive talent like data scientists and AI/ML experts. Given a scenario where projects still need to be completed, but resources are not available or accessible, how can enterprises continue to grow and expand their AI/ML projects? The answer lies in leveraging automation to empower an entirely new class of AI/ML developers within existing BI organizations. AutoML 2.0 platforms can accelerate nearly all of the steps required in developing AI/ML solutions and can provide a two-fold benefit to enterprise businesses: First, by making the AI/ML development lifecycle easier, AutoML 2.0 can accelerate AI/ML project timelines from multiple months to just a few days. Second, and even more critical, investments in AutoML 2.0 platforms can empower an entirely new class of users: Business Intelligence developers and Data Engineers. These new AI/ML experts, armed with AutoML 2.0, can help the organization scale AI/ML investments not just during a downturn, but can also provide an easily accessible resource as the economy stabilizes and returns to growth.
Investing for a Recovery
Regardless of the length of our upcoming recession, its effects are likely to be felt by everyone and to be significant. During these economic slowdowns, the gut-level response might be to reduce investments in projects that leverage AI/ML technologies because of the long timelines required before a return can be measured. The real problem that enterprises must solve, however, is not whether AI/ML investments are valuable. Instead, the focus must shift to enabling a larger class of users already inside each enterprise by adopting new technologies like AutoML 2.0 to help create a more empowered technology team made up of Business Intelligence professionals and Data Engineers. These existing and plentiful members of the organization can help businesses bring their current AI/ML projects to fruition, while exponentially scaling the organization’s ability to respond during an eventual recovery when our world, finally, comes back together.
