We live in an age characterised by the need to move fast, to learn from our mistakes, adapt and respond to change and predict future outcomes, so that we can gain competitive edge. Yet the pace at which business moves today and the availability of information means organisations are expected to make decisions and deliver change at a rate unheard of just a few years ago. The problem? We as humans don’t have the processing power to accept all available inputs, apply the lessons we learn about what does and doesn’t work quickly enough to drive incremental improvements, let alone big, disruptive innovations.
Most current prediction models have their roots in statistics innovations dating back to the 19th and 20th centuries. They’re often manual, usually painstaking and always very time-consuming. In those days, however, we had limited access to data – slow processes didn’t matter when there wasn’t much information to sort. Whereas today, huge amounts of data is generated each day and this is only increasing thanks to IoT, social media and online interactions. As a result, these dated approaches are no longer effective. We must look to other technologies and processes to support us.
Rise of the machines
Enter the rise of machine learning. Why? Because machines can learn at speed and on the go, aided by the data they’re processing. They can apply lessons learnt or insights captured from past behaviour to improve future performance. What’s more, they can perform these tasks without human intervention.
Machine learning essentially means that computers can improve over time at a particular task without being programmed. There are two benefits to that. First, you don’t need to redesign your programme every time the process changes slightly. Second, computers can now begin to extract insight from data and solve complex business problems which human judgement increasingly cannot manage. The machine is no longer constrained by the smartest human available to write code for it.
Whereas before machines would be used to complete a few tasks within a workflow, they are now close to executing the entire process, with humans only required to fill in the gaps. This provides a raft of capabilities previously considered too complex to be realistic. For example, predicting customer churn, prescribing the right drug, or even something as complex as helping pharmaceuticals stockpile drugs in preparation for an anticipated outbreak.
It can also help businesses reap measurable benefits in specific vertical sectors. In fact, in its report on the subject, “An Executive Guide to Machine Learning”, McKinsey states that European banks have replaced statistical modeling approaches with machine learning techniques and gone on to see improvements between 10 and 20 per cent in new sales or reduced churns.
Employing intelligent algorithms in company processes also gives skilled employees access to data-driven insights, which in turn can enable them to generate new opportunities for the business. When humans and machines work together in this way, the possibilities are huge – a more informed, data-savvy workforce will be better equipped to identify problems and opportunities, and then decide the right action in response. What’s more, with machines taking care of the number crunching, IT departments and business functions also have more time to be creative and deliver the innovations needed to improve performance.
Machine fuel = good quality, reliable data
Machine learning is becoming such a big part of our lives, that often we don’t even recognize all the ways it is being used by applications to deliver services. The chances are that if you’re using a navigation service while driving a car, reviewing the weather or using speech recognition on your smartphone, machine learning is in play and it’s improving the service each time it’s delivered.
Yet as with all technologies and new techniques, there is the potential for dangerous malfunctioning. The data fueling machine learning algorithms must be reliable at all times. If data isn’t clean and trusted prior to its analysis, it could cause misleading insights or even dangerous actions, especially in the case of drone deliveries or driverless cars. As a result, organisations must ensure data quality to guarantee the accuracy of their fast-responding machine systems – a glitch could be fatal.
CIOs take the helm of data management
Without clean and trusted data, machine learning won’t be fruitful. It’s time for CIOs to take the helm of data driven initiatives so that the business can benefit from fast and accurate insight and action, with as low a risk factor as possible.
Making that happen with a trusted machine learning capability will require the support of an end-to-end data management strategy. This will ensure the access to troves of data from all sources and sizes and then make it useful in a clean, safe and connected fashion. By pairing machine learning techniques with data management hygiene and quality of available data, organisations will quickly be able to spot new opportunities to disrupt their markets, and have the tools in place to seize on them.
This article is from the CBROnline archive: some formatting and images may not be present.