Author: Branko Bugarski, General Manager, HPE Software, Hewlett Packard Enterprise South Pacific
One of the most common misconceptions about technology is that people will always be on the losing end of new advancements. It’s an easy assumption to make considering the number of past human occupations usurped by machine automation. But the replacement phenomenon isn’t linear. There are boundless examples of new technologies that have failed to approximate the value delivered by the humans they replaced.
It’s common to apply automation with too broad a brush, asking machines to do things that only humans do well. This includes tasks like answering telephones and reading facial expressions. Meanwhile, in other disciplines, we require humans to complete data-driven tasks that machines do quite well, such as deciding how best to arrange store inventory.
Thanks to big data analytics, we can correct both missteps. Rather than replacing humans in a one-for-one exchange, big data analytics can be added to human-led processes, creating a collaborative hybrid of human and machine. By adding more data at the right moment in a process, big data analytics narrows the situations in which humans must guess the right decisions to make.
By making it possible to apply automation far more judiciously, big data allows machines and people to collaborate on decisions about processes and policy, leveraging the strengths of each.
For global business, this is a huge advantage because technology can be added and replaced as needed to deliver on whatever intelligence automation the organisation needs with near-perfect reliability and continuity.
While large-scale big data analytics installations are already in place around the globe, they have been mostly out of public sight, leveraged for confidential, high-value purposes. In 2016 this will change, bringing the analysis of big data front and centre in a wide range of business applications.
Here are five critical business benefits we predict big data analytics will begin to deliver in 2016.
1. Optimisation of labour
We all have hunches and ‘gut feelings’ about what to do, what’s right, and what’s wrong. But intuition without hard data to back it up seldom leads to an ideal choice. Optimisation happens when data drives a decision and is supplemented with human intuition.
For example, drivers of commercial vehicles relied primarily on intuition and prior experience to guide their decisions about which route to take. When given telematics and route optimisation data, people can vastly improve their driving efficiency and use their intuition to problem solve when necessary. These types of process hybrids allow both machines and humans to be their best ‘selves’ and bring optimal value to business processes and customer experiences.
By supplementing human-led processes with more data-driven decision support, big data analytics enables a closed feedback loop that can optimise human processes.
2. Choice in a multichannel world
People have strong preferences about channel. Research shows, for example, that millennials will avoid processes that aren’t offered via mobile or social media, their ‘native’ channels.
Channel diversification is great for user choice, but it creates a challenge for businesses, which haven’t had the technology support to make every channel equal in experience. For example, call centre agents use decision support tools to help them resolve issues according to policy, but if the system data and policies aren’t identical to those in other channels, the customer experience splinters. This leads to user frustration and confusion because different channels may yield different results.
Big data analytics can help organisations become channel agnostic. When you’re able to analyse big data quickly and accurately, every channel can draw on the same data sources and policies, giving assurance that all channels work equally well. Moreover, big data analytics can support frictionless cross-channel processes, meaning employees and customers can always choose the channel that is most convenient at any given time
3. Orchestration of processes and policy management
In any business, policies and processes are inextricably linked. Processes must operate within the policies set by the business, and be reviewed periodically to ensure they are not out of date or hinder the business.
When processes and policies are implemented through technology, breakdowns in process-policy alignment become evident through business outcomes. This is a good thing because when a process results in an unexpected outcome, it creates an exception, which tells us exactly where process or policy improvements need to be made. For example, if a customer abandons an online shopping cart in favour of a service representative or call centre, this indicates an area ripe for improvement. Big data analytics provides the means to track and analyse these interdependencies, thus avoiding the problems they can cause.
In addition, big data analytics will help reduce the overall cost of business process operations. For instance, big data analytics can help organisations reduce exceptions that cause costly escalations by identifying situations that fall outside of automated process handling.
4. Automated personalisation
In the past, personalisation tasks being given to people made sense, as machines have not traditionally been very good at making subjective—what are often called ‘squishy’—decisions. These include human communication cues such as nonverbal behaviour, facial expressions, and tone of voice.
In 2016, companies will begin pushing the leading edge further in terms of allowing machines to simulate squishy data. Big data analytics makes this possible by assimilating vast amounts of information, including the types of data that were too slow and expensive to collect and analyse in the past, such as communications and case records for knowledge workers. As the machines get better at interpreting a variety of data types (so-called ‘unstructured’ data) and collating it with vast quantities of structured data, they can begin to improve and accelerate both employee-owned business processes and customer-facing experiences.
Machines will begin to replicate human decision making, which human operators can oversee and deliver. This simulation with big data analytics can be enabled by better instrumentation and linkage between all machines and humans.
5. Greater instrumentation of dataRead more: The week in security: Apple-FBI encryption stoush dominates RSA; Aussie execs less hands-on than APAC peers
The value of a data-driven decision depends on the quality and quantity of the data and being able to deliver the insight to the right decision maker at the right time. This is about thinking more broadly and gaining a better understanding of how everything is interconnected. To collect intelligence from as-yet-untapped data sources, organizations will need to implement more precise instrumentation of the actions of and interactions between humans and machines.
For humans, it means collecting and analyzing clickstreams on everything we do online, from shopping to order entries on the job. It can also involve offline data, such as when we order takeout or how we navigate while driving. For machines, it means adding more sensors and meters to large capital equipment—everything from jet engines to cameras—to provide a richer set of data to be mined. In both cases, more data can lead to a better understanding of actions and behavior—as well as the domino effect that behavior has with respect to policies and processes.
As we begin to trap and utilize this data, we will begin adding another “sense” to our big data systems, enabling new types of intelligence that will create downstream innovation.
Creating a big data analytics strategy
The improvement of technology is, of course, an incremental process. But for business leaders, the lesson is not to wait for a perfect opportunity to leverage big data analytics and advanced automation. Instead, businesses should tap on opportunities throughout your organisation where big data can supplement human processes.
This can be achieved by looking for places to embed intelligence into human and machine workflows, finding areas to improve their interactions with each other. To succeed, businesses should also pay careful attention to the outcomes of such endeavors to make sure an improvement to one aspect does not mask unintended consequences elsewhere.
Big data analytics is set to give rise to distributed intelligence that can improve most anything humans do, but without a thunderclap of change. By taking a holistic view of integrated intelligence, you will be primed to reap all the advantages of our future in 2016.