Machine learning is having its moment. Almost overnight, it seems, the technology has sprung from relative obscurity and found its way into real-world implementations in almost every industry. Machine learning algorithms are helping smart systems teach themselves to be smarter, and are finding applications in everything from finance to engineering. Data-fuelled apps, meanwhile are transforming the customer experience with a new generation of real-time, context-aware services.
This has been a quite cathartic change for me. I have been championing machine learning throughout my career and can still remember the time a decade ago when I asked an engineering team to implement a machine learning model with over 1000 decision trees. You can imagine the strange looks I received.
In fact, we’ve reached a point where high-performance companies are looking at how they can scale machine learning across their operations; and companies which do not use advanced analytics, machine learning and AI capabilities are increasingly finding themselves at a competitive disadvantage.
In this blog series, I’m going to look at what machine learning means for business, and how organisations need to rebuild their approaches to people, processes and technology to realise the benefits of the technology. But first, I want to ask the question: Why is machine learning taking off now? For me, there are three key drivers:
The way businesses make decisions has changed. Decision making at businesses used to be on a grand scale: strategic decisions about where to open a new operation, for example, or whether to launch a new product range. Today, businesses are realising that money is made less in large-scale decision making and more in the day to day decisions they make one employee or customer at a time. These smaller decisions can be based on richer data where the feedback loop is much shorter and are therefore incredibly well suited to automation. As a result, machine learning algorithms are fast becoming key decision makers within organisations—deciding on everything from programmatic marketing campaigns to whether a financial transaction is fraudulent or not.
The customer experience is evolving—again. Over the past few years, we’ve seen clients increasingly interested in how they can transform the customer journey—from onboarding right through to retention. The digital revolution has offered new ways to engage with the customer and bring new levels of convenience and consistency to their experiences. However, these experiences have still been "dumb" and businesses are increasingly looking at how automated intelligence can be brought into the customer journey. This is the "intelligent" customer journey—the use of machine learning by companies to better personalise customer journeys, optimise products and services, and turn customer interactions into real-time living experiences.
Data continues to proliferate. Machine learning applications are only as good as the data that underpins them. As the digital revolution has progressed, the volume of available data has grown exponentially, providing businesses with a huge wealth of data to plug into their machines—data quality control processes notwithstanding. At the same time, big data analytics and cloud computing have matured, finally making it possible for businesses to ingest and process massive, complex and unstructured data sets in real-time and at marginal costs.
These three factors represent a perfect storm in which machine learning is prospering. But how can firms go about ensuring they deploy the technology in the right way? Tune in to the next blog, where I will discuss one of the most important elements to machine learning success: the human workforce.