As companies embark on their new digital journeys, the ability to understand and appreciate user behavior and needs is increasingly gaining importance. A key component is in leveraging data to build a better understanding of users, process context, make meaningful decisions, and execute upon those decisions.
One way to gather that knowledge would be through machine learning capabilities, which will soon dictate whether companies outstrip their competitors or scramble to keep up.
While the impacts of machine learning are most apparent in cool gizmos like the driverless car or through Siri, the most powerful examples are those in which the role of machine intelligence is so well-integrated that it seems almost natural. This is seen in supply chains when inventory is replenished at just the right time or in proactive maintenance when failures are intercepted before they can cause damage. There is no significant industry sector that hasn’t been affected so far — and many will undergo dramatic transformation as capability and creative integration unfold.
Machine Learning Goes Mainstream
When Bill Gates was asked in a Reddit “ask me anything” session what he would focus on if he were a current computer science student, he said, “The ultimate is computers that learn. … It has already made a big difference in video and audio recognition.”
While artificial intelligence and machine learning are not new concepts, they only recently have started making a significant impact and moved from esoteric jargon to mainstream conversations. Trends indicate that momentum will continue for a few reasons:
For one, there has been an explosion of readily available data through multiple channels. Now, we create as much information in two days as we did from the dawn of civilization through 2003.
Plus, Moore’s Law has prolonged its relevance more than any domain expert anticipated, leading to storage infrastructure and computational power becoming increasingly inexpensive. Today’s smartphones are millions of times more powerful than all of NASA’s combined computing in 1969.
Finally, the iterations over the 1980s and 1990s with improving infrastructure and availability of data have perfected the algorithms. Now, much of this algorithmic capability is available in public domain for research and in the form of open-source software.
We have seen the emergence of a new wave of companies for which core machine learning capabilities are fundamental to their market leadership. For example, Google can understand what users really want and match user profiles with relevant product advertisements to create an extremely profitable business model. Likewise, Amazon service is based on better appreciation of user needs and the ability to fine-tune logistics for both speed and cost efficiency. Uber combined the best of current technologies to solve a common consumer pain, giving rise to the “shared economy” model.
Each of these innovators have leveraged machine learning to reinvent the business process and have upended the prevalent business paradigm.
Implications for Building Future-Ready Businesses
While clearly there is tremendous power to be leveraged through data and algorithmic capability, we need to recognize that this is just a tool that can be used to enable businesses or we will end up creating a huge overhead in terms of systems and process.
The technologies that enable machine learning are increasingly commoditized, while the real value is in developing complex algorithms around creative integration to solve current business process challenges. The future will see a seamless integration of machine intelligence into the workforce. Here are three ways businesses can prepare for that future:
- Put Strategy First
Machine learning augments good strategy, hence the initiative should come from top executives defining which business areas and problems to focus on. This could be framed in terms of understanding user needs better in a customer-facing process, predicting machine performance and inventory movements in a supply chain, or identifying risks through unusual human or system behavior in a financial environment. Success needs to be framed in terms of how existing business process can be changed by having an improved intelligence.
- Contextualize the Solution
Companies must integrate machine learning into their processes to transform business problems into user solutions. Make sure that it combines seamlessly from the perspective of the end user so that each end user can give feedback, enabling continuous improvement.
- Change the Mindset
Machine learning is more than just another layer on top of existing processes: It represents a fundamental change in the way companies approach their problems with data and experimentation, replacing instinct and tradition. This requires a cultural change in the workforce as organizations revamp their information flow and relook at existing performance metrics to better enable the transition.
We are on the cusp of a massive revolution that will affect every aspect of both our business and our personal lives. The companies that recognize opportunities to use machine learning today will be the ones leading the pack tomorrow.
Image Credit: CC by Andrew Kuznetsov