CIOs getting to grips with machine learning and cutting through the hype
What do you need to do to become a first-mover CIO when it comes to machine learning? Developing methods to monitor machine-made mistakes and redefining job descriptions to prepare for the onslaught are key, according to a recent report from ServiceNow.
The study, titled ‘The Global CIO Point of View’, which polled 500 CIOs across 11 countries in 25 industries, argues that CIOs are adopting machine learning and ‘cutting through the hype.’ The use of machine learning is a major factor in driving through the new CIO agenda, as the research puts it: driving digital transformation, redesigning business processes, expanding the skills of the workforce, addressing business challenges, and competing to lead through improving data quality, focusing on customer experience, and more.
The first aspect to consider is whether the CIO is leading the charge. Almost three quarters (72%) of those polled said they were leading their company’s digitalisation efforts, with more than half (52%) adding machine learning was a vital part of it. Yet it is a team process; only just over a quarter (27%) say they have hired employees with skill sets focused on working with intelligent machines.
For those who are setting the pace in bringing machine learning to their organisations, they have confidence in their designs. Almost 90% of first-movers polled say they expect automation to support top-line growth, compared with only 67% of others, while it is a similar story when it comes to developing a roadmap for future business process changes (70% first-mover, 33% other).
Part of this is down to an ambitious set of tasks organisations want the technology to accomplish. 68% say they rank automation of repetitive tasks as an important capability of machine learning, while 40% say they feel it is important to recognise data patterns. These are relatively simple goals; yet at the other end of the scale, more than half (54%) say they want machine learning to make complex decisions for them.
For instance, in retail, inventory can be monitored, then restocks can take place based on industry levels, and goods can be further ordered combining the two based on predictive analytics. For healthcare, the medical records and literature can be digested, treatment options suggested, and individual treatment plans devised.
“First-mover CIOs who combine machine learning with new business processes and skillsets will better support their enterprise growth,” said Chris Bedi, ServiceNow CIO. “They report higher levels of maturity in the use of leading platforms, which allows them to concentrate on innovation, such as automating complex decision-making, which immediately impacts the bottom line.”
ServiceNow recommends five steps for organisations looking to achieve value from machine learning:
- Build the foundation and improve data quality: Get the foundation right – capturing the right data to feed machine learning algorithms, which could also be data from outside the organisation – and utilise the right technologies to simplify data maintenance and the transition to machine learning
- Prioritise based on value realisation: In other words, don’t just lift and shift current processes into a new model
- Build an exceptional customer experience: Don’t think in terms of individual customers, but the entire customer journey
- Attract new skills and double down on culture: Skill sets for machine learning are many, and include engineering, mathematics, data science, and critical thinking. Don’t forget either about how employees may react to the shift towards working with machines
- Measure and report: CIOs “must set expectations, develop success metrics prior to implementation, and build a sound business case in order to acquire and maintain the requisite funding,” the report notes
You can read the full report here (pdf).
- » ServiceNow notes importance of chief human resources officer in digital transformation tangle
- » A guide: How to better manage your personal information
- » Gartner identifies seven steps to bring down risk of security threats from Spectre and Meltdown
- » Three key reasons why apps and network teams should talk more
- » Lack of knowledge around biometrics apparent – but could they be coming to a workplace near you?