Senior lecturer in operations management at Bradford University School of Management, Dr Olga Matthias, considers the implications of big data for business and looks at whether this next generation in analysis will lead to better business decisions and results.
The opportunities and challenges of big data for businesses are the subject of furious debate. There are enormous social and organisational challenges as a result of this exponential technological growth and businesses continue to discuss whether the usage of large amounts of data leads to better decisions and results.
Gartner (2014) has said Big Data is re-defining the relationship between human and machine but as yet we still have the unfulfilled promise of business productivity. What effect does and will big data have on the business as you try to manage day-to-day operations, your staff and customer requirements?
As yet, there is not a universally confirmed definition of big data. Some limit big data to digital inputs, such as social network interactions or analytics logs (Taylor 2012). Others include traditional data sets too (Arthur 2013). There seems to be agreement that anything that cannot be processed using traditional database management tools and applications is big data.
Essentially big data is a large data set that requires a new approach and makes full use of the increasing power of modern computers and storage systems.
Getting business insights from big data
Globally, more people own a smartphone than a toothbrush (Harding 2013). This quirky but slightly depressing fact is remarkable, given that smartphones only became widely adopted in 2007 (Ozgur 2011).
Social media allows businesses to tap into the ‘conversation’ customers are having (Chen et al 2012). This can provide key business insights that have never before been available. The idea is that big data shadows and information trails of people, machines, commodities and even nature (for instance for weather forecasting) can reveal secrets to us that we only now have the power and prowess to uncover.
The hype says that as we gather more data, we will be able to make better decisions to resolve operational business problems. Organisations now capture more customer data than ever before without the storage and retrieval costs becoming prohibitive. But, unless you know what you want, big data is a source for confusion. You need to be absolutely clear about what you want to know to fully realise the benefits of big data.
Turning big data into business opportunities
However, we can’t seize business opportunities if all this information isn’t usable. Usability makes big data have value to someone, essentially making it big information – and it is this which is the most important thing.
The implication is that organisations need to fuse the information (ie data represented in a processable format), meta-information (annotations, processes, workflows, human knowledge, feedback and history of previous work, lessons learned, quality scores, collective intelligence) available or dynamically created and models (algorithms to identify and extract patterns of information from the existing data). But still we need to interpret the information and then acknowledge the valuable pieces of knowledge from it.
With big data, businesses have to be able to extract patterns so they can understand what is happening. Then you have to understand why. The data does not tell you why. You look for patterns, and for glitches in those patterns, to see what they imply.
Because you have so much data you have incontrovertible evidence – you have in effect taken a census. You are not extrapolating. Done correctly this allows for exciting decision-making. Done badly – there is a huge operational cost to the organisation, and a cost that is usually hidden. For instance, a kitchen manufacturer found they could predict the type of kitchen and ballpark spend by tracking which browsers were used – Mac OS and Google Chrome meant a modern, expensive kitchen whilst IE meant more traditional style and careful spend.
Realising digital opportunities
Beware the potential for misinterpretation or mis-correlation. The oft-quoted Target example of the father who found out his daughter was pregnant through their targeting of her newly-changed shopping habits and the Google flu data – where algorithms are modified by the service provider for a variety of reasons in the name of agility and flexibility.
What you’re actually looking for is reliability AND robustness. Once you have that, does it mean better customer satisfaction and profitability?
No, because there are a number of people-related challenges to be considered, and overcome, if the promise of a brave new management world that is digital is to be realised.
There are data access problems, utility problems a need for new data science skill-sets. There are problems with how we think, how we react and respond to things at any given time. For instance speed – is speed a good thing? Does it mean we can make better, more informed decisions for ourselves and our organisations? Or does it mean we become confused? How much data can we cope with to optimise what we do and how we do it?
Is our behaviour influenced by the big data we see or does what we do affect the big data?
What the enthusiasm around the big data idea seems to indicate is that we are seeking an augmented reality. In an idea world, what we want is the ability to exploit big data so that we can set new levels of performance and new agendas for business.
This can mean a step change in competences for some of the tasks we expect staff to do as a matter of course once the world of big data is embraced. For instance, GCHQ employs more than 100 dyslexic and dyspraxic ‘ neuro-diverse’ spies to harness their analytical skills in the fight against terror because they are generally skilled at deciphering facts from patterns or events.
Yet, in the past, as Richard Branson would no doubt attest if he were here, dyslexics struggled in the traditional business world and most have had to learning coping techniques and mechanisms.
Customers have changed too. They are mobile and in control of the relationship with your business.
They would like to interact with your business using the channel of their choice no matter if it is the agent, online, or phone channels such as voice, text, mobile web, smartphone apps or even social networks, at a time of their choice. They expect channels to be connected in such a way that conversations started on one channel can be continued on another.
It’s a brave new world, the world of Big Data