More analytics – speed and agility through prediction

The insights from data are only useful if they generate payback for the business or the consumer, preferably both.

All told, this means that when people talk about data – ‘big’ or ‘small’ – they should always be talking about data analytics as well.

Once large datasets are analysed, trends can be revealed, correlations can be determined, and businesses can run more efficiently.

The 6 V’s of data (volume, velocity, variety, veracity, vulnerability and value) are driving an increasingly pressing need for smarter and more automated analytics, more commonly referred to as ‘advanced analytics’.

A new analytics market is evolving, with specific characteristics, including:

  • Fast, automated analytics that informs operational decision-making in real-time, or near real-time
  • New predictive methodologies, such as machine learning, which drive recommendations that can keep pace with market behaviour
  • Analysis of huge data sets, from a variety of sources, through self-service tools
  • Findings delivered in an easily visualised format, which are consumed through a single inter-connected technology platform.

>See also: 10 trends that will influence analytics in 2017

Prediction, not reaction

The ‘science’ of predictive analytics is at the forefront of making sense of large data sets.

It provides a set of techniques that deal with extreme complexity, at speed and with agility, extracting meaningful information from present as well as historical data sets.

It spots patterns in data and predicts unknowns, future outcomes, and trends.

Traditional analytics insights and descriptions (what happened and why?) are evolving towards predictive and prescriptive applications (what will happen and how can we make it happen?).

This means far more tangible return on investment for organisations.

Cognitive analytics

Focusing on one branch of predictive analytics, the advent of cognitive analytics means we can now automate analytical thinking and self-learning through machine learning.

Learnings are fed back into the analytics ecosystem to be applied in future situations, to answer new or related questions. Every time the mechanism becomes smarter.

Needless to say, analytics at this level has a vast range of applications. Fraud agencies will be able to develop risk engines that flag candidates for further examination.

Retailers will be able to optimise decision processes such as monitoring and addressing inventory levels, or adjusting in-store pricing in real-time for flash sales.

Learning algorithms are increasingly anticipating what we want, delivering recommendations and offers to the consumer based on what they’re about to do, and not what they’ve previously done.

This anticipatory analytics is enabled by access to vast new stores of data, from customer behaviours and external sources, which enable organisations to deploy ‘situational selling’ based on the specific individual and their situation at that moment in time.

>See also: Bridging the business intelligence and analytics gaps

Some of the world’s top companies continue to invest heavily in Big Data, machine learning, and artificial intelligence capabilities.

Over the next few years, we can expect to see significant advances in machine learning, reducing the time to insight and manual data wrangling required.

This will then free up significant chunks of time for employees to focus on driving efficiencies and work on directly revenue-generating activities.

Data visualisation

Advanced analytics may be powerful and impressive, but the data and findings must be presented in an accessible, readable format which makes them clear.

Developments in data visualisation technology now allow users to view and manipulate hundreds of variables and parameters, and the impact these are having has led some commentators to suggest that ‘visualisation’ should be included as another of the data V’s.

New opportunities

The powerful combination of behavioural data and predictive analysis is opening up new areas of opportunity for analytics businesses:

  • Dynamic pricing – from ride-sharing services to sporting events, this real-time situational pricing capability allows companies to balance supply and demand. Research has shown that consumers are becoming more accepting of these practices, as long as the variation remains reasonable.
  • Pre-emptive safety – the ability to predict events such as fires or failing machine parts.
  • Rehabilitation assignment decisions on an individual offender basis, drawing on predictions of future repeat offences.

 In addition, complex event processing and streaming analytics software can detect poor user experiences, enabling brands to offer proactive customer service. A customer who presses the start button of a new device five times in a row would indicate they’re having trouble operating the device, and could be offered real-time support or an instructional video via their mobile device.

While this may all sound like the human element of analytics will be circumvented, this won’t be the case at all.

>See also: The 3 pillars of big data analytics potential

Models must be built and tested, designed to accommodate incomplete information, and developed to adapt to mistakes and failures. Actual and expected performance must be monitored and compared, and any disparity addressed.

Above all, transparency of advanced analytics will still be demanded for the foreseeable future so that organisations can track what is happening and why.

With people in every area of business often struggling with data and analytics, the opportunity for analytics companies is huge.

As data sets grow, we’ll see predictive algorithms used more and more as a methodology for mining data where traditional techniques can’t be used.

Predictive analytics will continue to evolve from an innovative extra capability to must-have functionality. There will also be experience the integration of advanced analytics, predictive analytics and machine learning.

As the complexity in the analytics world continues to increase driven by the wider adoption of digital technologies and changes in consumer behaviour, the focus will continue to be on developing solutions that manage the complexity of analytics engineering and statistical modelling to deliver easily consumable products and platforms for businesses.


Sourced by Rafa Garcia Navarro, ‎chief analytics officer UK&I at Experian

Avatar photo

Nick Ismail

Nick Ismail is a former editor for Information Age (from 2018 to 2022) before moving on to become Global Head of Brand Journalism at HCLTech. He has a particular interest in smart technologies, AI and...

Related Topics

Big Data