It has the potential to be big business. According to a recent study by the research group CEBR, big data could contribute £216 billion to the UK economy between 2012 and 2017. To put that into context, this is more than twice the annual cost of the NHS or 22 per cent of the UK’s national debt.
Businesses of all sizes have been using data analytics to look at how well they perform – and to seek out ways of doing better – for years.
Supermarket chain Tesco has built up a vast database through its Clubcard loyalty scheme and used it to drive the growth of its business over more than a decade. Smaller businesses have also used data analytics successfully with nothing more complex than a couple of computer spreadsheets to identify new sales opportunities or to match supply with demand.
The current buzz around big data, though, suggests taking analytics to an altogether different level. When academics and experts talk about big data, they do sometimes mean huge, purpose-built databases in the Tesco mould.
But they are also talking about tapping into a far, far greater wealth of information, much of which is captured by the internet, or held in so-called “unstructured” data formats, including presentations, documents, video files or e-mail conversations.
Then there are the vast quantities of data captured every day by social networking sites, such as Facebook and Twitter, much of which is available to businesses wanting to add trend information to their own predications of the future.
Harnessing this data is increasingly important to staying competitive and helping companies grow. “It is fundamental,” says Peter Lumley, an analytics expert at PA Consulting. “It is about having a fact base to base growth on. You want to know who you are dealing with and which customers make the right profit margins.”
This is made easier by the coming together of cheaper technology and data sources. The tools for carrying out even quite sophisticated analysis might cost as little as £200 for a single user or a few thousand for a developer tool. And businesses may already have the data, Mr Lumley points out.
The driver to invest in analytics technology is often the fierce competition some businesses face. Retail is an early adopter of the technology partially because retailers have access to plenty of data, but also because of the cut-throat nature of the high street, especially over the last few years.
“The business-to-consumer sector is very competitive at the moment; firms are competing to win new customers, so analytics is being heavily applied to be more accurate in their view of the consumer,” says Ray Eitel-Porter, who heads the UK analytics business for consulting firm Accenture.
The beauty is you don’t have to be a Tesco to create value from big data – almost any company can do it
He estimates that companies using big-data technology can see a 20 per cent improvement in conversion rates when it comes to winning new customers. “This is because the offers are more tailored and more relevant,” he says.
This can be as simple as offering customers complementary products for their shopping baskets – say batteries for electronic toys – or more complex deals in areas such as mobile-phone contracts and financial services.
Online retailers, of course, have long offered such recommendations. But data-driven approaches to increasing customers’ spending are moving into more complicated areas than books and CDs.
“Customer experience is joining up what you do in the front and back office and the supply chain,” says Eddie Short, head of business intelligence at KPMG. “But even sophisticated retailers have a long way to go to match the likes of Amazon in terms of being data driven.”
This, in turn, is prompting companies to use evermore complex algorithms to forecast customer behaviour, and work out where and when to target their sales and marketing efforts.
The science behind this branch of “predictive analytics” has origins far from the high street: some techniques were developed originally for shooting down hostile missiles. “Companies use this to understand how behaviour changes as the consumer nears the point of purchase,” says Accenture’s Mr Eitel-Porter.
Another example is technology developed by Featurespace, a UK company spun out of Cambridge University. Developed originally to detect fraud in the online betting market, its technology can look at normal customer behaviour and detect anomalies, says chief executive Martina King.
And, according to commercial director Matt Mills, it allows marketers to respond much more quickly to changing events. “It gives marketers a different way to think about targeting,” he says. “Most organisations, even if they use big data, are still making assumptions. People look at the historic data and make assumptions about the next steps. That works wonderfully if things stay the same, not when they change very rapidly.”
As a result, companies selling directly to consumers are adopting new technologies to deal with one of the downsides of the big data revolution: the rapid swings in public sentiment that are amplified by social media sites. But advanced analytics are also being brought into service to help businesses cope with anything from changing weather patterns to swings in demand for travel.
“The beauty is you don’t have to be a Tesco to create value from big data – almost any company can do it. The technology infrastructure is really cheap and can be put together rapidly. In the past, data had to be very well structured to make sense of it, but now it can be unstructured – a lot of data is just available,” says Ralf Dreischmeier, global leader of the technology practice at consulting firm BCG.
He adds that big data is being used to spot new business opportunities, create new products, or even develop cross-business or cross-industry collaboration, such as between telecoms companies and retailers, to build a more detailed picture of where their customers are, what they are doing and what they might do next.
But despite the publicity, and even hype, surrounding big data, experts suggest it is a tool that is here to stay.
“It’s a fundamental change,” says Mr Eitel-Porter. “People ask if it is a fad from consultants or academics. But we are living in a digital world. Everything will be captured in a digital form and it can be analysed.”
HOTELS
DATA WITH STAYING POWER
Like many companies in the hospitality and travel business, hotel operator Jurys Inn had relied heavily on search-engine advertising to win online business.
But search engines, according to Jurys Inn’s online marketing manager Carol Walker, usually work best for attracting guests who already know about the brand or even a specific hotel.
Conventional print advertising, on the other hand, raises brand awareness, but is not flexible enough to reflect changes in hotels’ occupancy rates or shorter-term movements in supply and demand.
The hotel chain needed a way of matching its advertising campaigns to its vacancies, but also to attract potential guests who might not have stayed at an individual hotel, or a Jurys Inn, before.
Search ads are short and to the point, while display advertising allows the hotel to include a lot more information about rooms, conference facilities or local attractions, for example. What the company needed, according to Ms Walker, was a way of “turning advertising on or off for individual properties”.
By working with Agenda21, a specialist digital agency, Jurys Inn is able to combine data from its internal reservations and revenue management systems with an up-to-date “inventory” of online advertising. Data analytics ensures that the hotelier focuses its advertising bookings on properties that have vacancies and, at the same time, the adverts contain enough information to ensure that new guests try out the brand.
“Search advertising is important at the point of purchase, but with display we are trying to introduce people to the brand, people who might not be looking for a Jurys Inn property but a hotel in London, Edinburgh or Glasgow, and we can show them the hotel in more detail,” says Ms Walker.
“With the granularity of data we have, we can now see where we need to increase or reduce advertising investment. If we know a property has good availability, we want to make sure we are doing all we can to fill it.”
CONSTRUCTION
A TALE OF HOUSE-BUILDING
Story Homes is a fast-growing company, which expects to increase turnover from £50 million last year to £500 million in 2015. This year, it will complete between 250 and 275 homes, against 152 in 2012.
The company, which sells higher-end, new-build homes in northern England and Scotland, has funding in place to expand, and a sizeable land bank of sites it owns but has yet to develop. As part of its growth plans, the company is using data analytics, both to make sure it is buying and developing the right plots, and putting the best-selling house types on them.
Using data analytics is a key part of ensuring the company’s growth plan is successful, says managing director Steve Errington. “A key aspect was to put in a planning system to manage the business,” he says.
The system, based on Anaplan analytics software, allows the company to build a four-year business plan, down to the level of an individual house plot. The company uses this to track sales across the business, to adjust the type of plots it is looking for and even, potentially, the type of house it builds.
“Our land bank is in excess of our current delivery rates, which is why we’ve been able to grow so quickly,” says Mr Errington. “But before we put in the planning tool, we were not in a position to do quick reporting of ‘what if’ scenarios. We had identified lots of land opportunities, without a strategy.”
The company is also putting feedback from home buyers into the system, based on their likes and dislikes about the Story Homes properties they have bought. This is then passed on to the company’s architects and designers to help them refine the type of homes being built.
“We build a higher quality product, at a higher price, than the big boys,” says Mr Errington. “Our model only works if we can get a price premium, to justify our higher land and building costs.”
