Analytical acumen of greater value than Big Data

8th September 2017 By: Schalk Burger - Creamer Media Senior Deputy Editor

The acme of analytics does not depend on the size of the dataset but on the capabilities of analysts to derive meaning and, consequently, value from the data, says Barclays Africa chief data officer Pieter Vorster.

Therefore, so-called Big Data analytics does not differ from typical data analytics, he says, advising that companies adopt analytics best practice before trying to jump into Big Data analytics.

“Some of the smartest insights we produced were derived from 1 GB of data and the use of regression analysis, a technique that was developed in the eighteenth century.”

Technical problems, such as effective data sourcing and quality management, must be overcome to provide business insights. Bad-quality data will lead to bad-quality decisions.

“Advanced analytics is the ability to use technology to solve complex pieces of information and gain insights more rapidly. These tools provide the opportunity to derive information from disparate pieces of data.”

Technology does not replace or remove the need for high-quality data and good data sourcing systems underpinned by good data management and analytics principles. The ability of companies to use analytics systems to compete correlates with good-quality data and the speed at which they can use the data.

Rather than viewing analytics systems only as supportive technologies, companies should view analytics projects as journeys to make business data valuable.

“Start with a business problem that can be addressed by using analytics. For example, descriptive analytics can help a company pinpoint the specific problem that needs to be solved,” says Vorster.

Barclays Africa’s approach to the new business technologies, including artificial intelligence and data analytics, is to focus on making its business smarter, instil the foundational competences first and build systems that are scalable, he adds.

Building the data sourcing, quality and governance systems to support analytics systems and developing the competences to create products and services should also be done alongside a fail-fast, iterative innovation process.

Many of the systems and ideas, especially during the initial phases of Big Data projects, will not work. Companies should, therefore expect failures and move on while noting the lessons learned. However, the focus should always be how the company can solve business problems by using the appropriate technologies and the right skills and people.

Barclays Africa garnered an award from market research firm Gartner for its data analytics projects, not because they are done at massive scale, but owing to Barclays Africa applying the principles of data analytics across the bank, emphasises Vorster.

Meanwhile, while a lack of analytics skills is common in most industries, often the best analysts and developers of analytics systems are those with curiosity and passion, and not necessarily tertiary degrees.

“You do not need a person with a doctorate to develop these systems. A person with passion and curiosity will experiment, study during his or her own time and familiarise himself or herself with the systems and diverse skills required. Additionally, much of the materials are available free of charge from online training platforms.”