In 2017, analyst firm IDC made a shock prediction that the world would be producing 163 zettabytes of data a year by 2025. To put that in context: in 2017 the world produced 16 zettabytes of data, which is already more than all the data generated in the past 5 000 years of human development. This data deluge is inevitably going to pose challenges for businesses in how they collect, store, analyse and derive value from their data. And time is running short for practical solutions to how we manage this huge volume of data: the same IDC study predicts that by 2025, 20% of all stored data in the world will be critical to our daily lives.
For C-level business decision-makers, the big challenge is how to leverage their company data to produce insights that can inform better sales and marketing practices, improve customer relationships, optimise production and increase efficiencies within the supply chain. In this, predictive analytics driven by machine learning algorithms are emerging as the go-to tool for executives. But this is not without its perils.
Bridging the millennial gap
One of the major drivers of advances in modern analytics is the meteoric rise of the data science profession, which despite its relative youth holds huge influence over how companies mine their data for business value. Demand for data scientists is set to soar by 28% by 2020, according to IBM, with 59% of the demand originating in the finance, insurance, professional services and IT fields. Data science as it is applied today is a relatively new profession, and many of the new skills entering the workplace lack real-world business experience as they’re most likely straight out of university and placed into key roles due to a prevailing data science skills shortage.
This is causing a disconnect between the analytics teams and the C-suite, especially for the telecommunications, retail and financial services industries, where most of the executives are not as up-to-speed with new technologies as their younger peers. This means they are often left at the mercy of younger data scientists who don’t always have the depth of industry and business knowledge needed to produce analytical insights within the correct context. There’s no question the younger data scientists are skilled – often they develop near-perfect algorithms that can produce insights within a narrow field – but many businesses are so vast and require insights into such a dizzying array of touch points and lines of business that single-use algorithms often have little value to the overall business performance.
Delivering value to the C-suite
As a general business rule, CEOs, CFOs and members of the board don’t choose to implement changes or update processes simply because it seems like a good idea. If the proposed implementation – whether it’s a new process, piece of technology, organisational change or something else – doesn’t reduce risk, cut costs, or bring benefits to the broader organisation, chances are they’ll dismiss it until it can meet such criteria. No algorithm is going to replace a board of directors, no matter how smart it is. What business analysts and technology vendors need to focus on is utilising new tech – for example an advanced algorithm – to deliver critical insights in a format they understand and can use to make better decisions.
Working with global vendors offers distinct advantages in this regard: SAP Analytics Cloud, for example, takes data, matches it to predetermined KPIs, and applies machine learning capabilities to visualise the data in the most appropriate way. Business analysts then have the option to adjust the visualisation before it is delivered to the board as live, accurate insights that executives can use to improve performance across the business.
The deeper level of value delivered by the machine learning capabilities is tracking and measuring how one KPI – for example, improved efficiency within the supply chain – influences and changes the other KPIs, for example sales volume in a particular store or region. It’s easy to think you need a bigger sales base when your sales start to slow down, but it might be a simple costing issue that can be resolved through dynamic pricing.
But it’s unlikely companies will unearth this type of value if they rely on bespoke single-use algorithms designed by data scientists who may not have the business experience needed to connect the various dots between disparate lines-of-business.
The only way to bridge this gap is through analytical platforms that deliver automation capabilities as well as a host of toolsets that can be applied to produce deep analytical insights and data visualisation in formats that the C-suite can easily understand and derive value from. SAP’s answer to this is SAP Leonardo, a digital innovation platform that combines emerging technologies such as IoT, big data, machine learning, blockchain and predictive analytics on the SAP Cloud Platform to deliver all the tools that organisations need for digital transformation.
