Preparing for prediction

Data is more than simply a matter of record. It is becoming more widely recognised as a valuable source of business intelligence. Recognition is growing and an increasingly wide range of internal and external data sources are available. These factors, when coupled with the latest analytical technology, can significantly enhance business processes.

Analysis has typically been about looking in the rear view mirror to understand what has happened but advances in technology are enabling a shift in intelligence from hindsight to foresight, known as predictive analytics. 

By definition, predictive analytics are a set of statistical tools and technologies that use current and historic data to predict likely future behaviour. They can help businesses understand their problems and formulate a strategy that can generate significant business value and drive competitive advantage. For example, a lender identifying potentially bad customers, an insurer looking to spot possible fraudulent activity, or a marketing function trying to tailor campaigns to accurately target potential customers.

Organisations that understand the value of their data will also recognise that it is an asset that requires protection and, in parallel with exploiting the data, they will develop suitable approaches to ensure that data is secure and used in appropriate ways.

Before predictive analytics can be deployed into an enterprise, some fundamental questions about what the organisation wants and its ability to achieve those goals will need to be answered. Predictive analytics can offer substantial and measurable business benefits and these should be captured in a business case that is properly aligned to the organisation’s strategy and reflects an understanding of how it will impact revenues and assist in the elimination of risk.

Then, a root-to-branch review of current information management capability is needed. Without a clear understanding of where you are it is far harder to define – let alone achieve – the desired future state.

Analytics should be viewed as an ongoing undertaking in order that a sustainable capability is developed that supports the overall business strategy of an organisation.

When it comes to which software or hardware to use for predictive analytics the options available are expansive. There isn’t a one-size-fits-all recommendation for technology solutions as these decisions are influenced by industry type, budget and culture.

For example, when deciding whether you go for a cloud-based solution, a dedicated managed service or a buy-and-build solution, there are a number of issues that need to be considered. Capital expenditure for many companies today is still being squeezed, so solutions that offer ‘smoother’ operational expenditure may be favourable. 

Day-to-day operations of the technology will need to be clearly defined. Exactly who will have access to information? Are legal requirements around keeping data being adhered to? How will suspected incidents be assessed and escalated and will monitoring be kept current and relevant?

The changing approach taken to extracting value from data, with advances in analytical technology and a broad array of delivery models offers organisations a powerful combination for value creation. Heady stuff, but getting the next generation of intelligence off the ground requires a firm foundation in the basics of information and performance management, paired with a business case that speaks the language of value.

Kate Russell

Kate Russell is an entrepreneur and the MD of Russell HR Consulting. She is an author and public speaker on all HR issues surrounding SMEs for future reference.

Related Topics