We are entering the era of the machine learning enterprise, in which this subset of artificial intelligence (AI) capabilities will revolutionize operating models, shake up staffing methods, upend business models, and potentially alter the nature of competition itself. The adoption of machine learning capabilities will be limited only by an organization’s ability to change – but not every company will be willing or able to make such a radical shift.
Very soon, the difference between the haves and the have-nots of machine learning will become clear. “The disruption over the next three to five years will be massive,” says Cliff Justice, principal in KPMG’s Innovation and Enterprise Solutions team. Companies hanging onto their legacy processes will struggle to compete with machine learning enterprises able to compete with a fraction of the resources and entirely new value propositions.
For those seeking to be on the right side of the disruption, a new survey, conducted by SAP and the Economist Intelligence Unit (EIU), offers a closer look at organizations we’ve identified as the Fast Learners of machine learning: those that are already seeing benefits from their implementations.
Machine learning is unlike traditional programmed software. Machine learning software actually gets better – autonomously and continuously – at executing tasks and business processes. This creates opportunities for deeper insight, non-linear growth, and levels of innovation previously unseen.
Given that, it’s not surprising that machine learning has evolved from hype to have-to-have for the enterprise in seemingly record time. According to the SAP/EIU survey, more than two-thirds of respondents (68%) are already experimenting with it. What’s more, many of these organizations are seeing significantly improved performance across the breadth of their operations as a result, and some are aiming to remake their businesses on the back of these singular, new capabilities.
So, what makes machine learning so disruptive? Based on our analysis of the survey data and our own research, we see four primary reasons:
1. It’s probabilistic, not programmed
Machine learning uses sophisticated algorithms to enable computers to “learn” from large amounts of data and take action based on data analysis rather than being explicitly programmed to do something. Put simply, the machine can learn from experience; coded software does not. “It operates more like a human does in terms of how it formulates its conclusions,” says Justice.
That means that machine learning will provide more than just a one-time improvement in process and productivity; those improvements will continue over time, remaking business processes and potentially creating new business models along the way.
2. It creates exponential efficiency
When companies integrate machine learning into business processes, they not only increase efficiency, they are able to scale up without a corresponding increase in overhead. If you get 5,000 loan applications one month and 20,000 the next month, it’s not a problem, says Sudir Jha, head of product management and strategy for Infosys; the machines can handle it.
3. It frees up capital – financial and human
Because machine learning can be used to automate any repetitive task, it enables companies to redeploy resources to areas that make the organization more competitive, says Justice. It also frees up the employees within an organization to perform higher-value, more rewarding work. That leads to reduced turnover and higher employee satisfaction. And studies show that happier employees lead to higher customer satisfaction and better business results.
4. It creates new opportunities
AI and machine learning can offer richer insight, deeper knowledge, and predictions that would not be possible otherwise. Machine learning can enable not only new processes, but entirely new business models or value propositions for customers – “opportunities that would not be possible with just human intelligence,” says Justice. “AI impacts the business model in a much more disruptive way than cloud or any other disruption we’ve seen in our lifetimes.”
Machine learning systems alone, however, will not transform the enterprise. The singular opportunities enabled by these capabilities will only occur for companies that dedicate themselves to making machine learning part of a larger digital transformation strategy. The results of the SAP/EIU survey explain the makeup of the evolving machine learning enterprise. We’ve identified key traits important to the success of these machine-learning leaders that can serve as a template for others as well as an overview of the outcomes they’re already seeing from their efforts.
Learn more and download the full study here.