Part 1 in the 4-part Predictive Analytics and Machine Learning series
Time and tide wait for no man, as it is said, and technological advancements will not wait for organizations to adopt them. Fear of failure, oddly enough, is often a reason for lack of tech adoption in the enterprise. Ironically, this reluctance is also a major reason that companies fall behind.
Every company looks for ways to differentiate in the market, and some grasp the value of new developments, concepts, and ideas faster than others. But sadly, while recognizing the potential, many companies never make it to the actual implementation stage. This is not a new phenomenon; it has been happening over decades as the tech world evolves.
Some decision-makers are masters of procrastination. Think how many CIOs have indicated in various surveys that data governance, data quality, a proper information strategy, mobile BI, or other similar topics were critically important and crucial to tackle within the “next 12 months.” Then nothing happens.
Learning without doing
Tech conferences continue to be well-attended – a good thing, since it is important for people to stay up-to-date on the newest tech trends and connect with peers to find out what they are doing and how. It’s not surprising that presentations with a current buzzword in the title enjoy large audiences, with topics such as “blockchain” and “machine learning” (ML) currently drawing the biggest crowds. While blockchain seems to be more difficult to grasp when it comes to broad use cases, ML is a treasure trove of interesting stories from virtually every industry and line of business. These applications and use cases should provide ample inspiration for those conference attendees, right? Well, not so fast.
In a Forrester study titled “Powering The Intelligent Enterprise With AI, Machine Learning, And Predictive Analytics,” over half of surveyed organizations identify concerns over privacy and compliance as a barrier for adopting predictive analytics and ML technologies. In fact, according to the survey, there are a dozen reasons for organizations not to engage in those new technologies, including GDPR and cost concerns, lack of skills and resources, and uncertainty about where to even start. Is that reason enough to not do anything? Is it reasonable to avoid going into uncharted territory because there might be a few bumps along the way?
Waiting for a sign
A survey conducted by IDC and SAP revealed that a mere 15% have adopted ML, 40% are planning to adopt ML with two years, 26% within two to five years, and 19% have no interest at this time. In other words, two-thirds are sitting on their hands, waiting for something to happen before they engage in an ML project or initiative. And a fifth of those surveyed are completely looking the other way. How does that align with everybody’s talk about transformation, differentiation, digitalization, even disruption?
Those interesting ML use cases typically don’t emerge out of thin air. Indeed, they require a lot of brainstorming, design thinking, or other favorite ways of strategizing about the intelligent use of information to gain a positive impact – from efficiency gains and quality improvements to risk mitigation, fraud detection, or entirely new business models. Waiting for that killer idea to miraculously fall into someone’s lap is not a good plan, especially in today’s tough competitive environment.
A very common question is “what are others doing?” That’s a good way to identify deployed technologies, selected vendors, and organizational structures. While it is a good approach to test the waters before engaging in a time- and money-consuming endeavor, it’s likely these decision-makers just don’t want to take a risk. As a result, they choose to copy and adapt someone else’s successful approach, even though differentiation went right through the window. Real transformation and innovation look very different across individual enterprises.
How to jump in: not feet first
To engage in analytics, Big Data, IoT, machine learning, or any other information-centric initiatives, organizations large or small need a solid strategy document that’s aligned with corporate goals. It should be written by experts from both the technical side as well as the various business functions. It should include aspects of not only the technical components and architecture, but also vetted requirements and a business case, priorities, organization and steering, sourcing, program management, education, support, and a roadmap. Otherwise, too much is left to chance with uncoordinated efforts that are often counterproductive.
Without an enforced strategy, mainstream initiatives, such as business intelligence, analytics, or data warehousing, are often undermined by rogue projects. That regularly leads to mistrust and confusion. At the same time, the interesting opportunities, such as IoT, artificial intelligence, machine learning, or predictive modeling, are often reduced to grassroots projects with little leverage because the bulk of the resources are spent just trying to keep the lights on.
How innovation happens: by not putting it off
Every time a market is disrupted or a particular company moves away from the pack, it’s because employees are encouraged to do some creative thinking. New technologies are evaluated by “first movers.” Risky projects with unclear outcomes get funded because there is a potential upside – and eventually turn out successful. Waiting for all the puzzle pieces to slowly fall into place, while a common practice in politics, is not a viable strategy when the markets demand transformation and innovation.
For an in-depth look into the intelligent possibilities for your business, review the August 2018 Forrester Consulting study, “Powering The Intelligent Enterprise With AI, Machine Learning, And Predictive Analytics,” commissioned by SAP.