Big Data has not yet led to big outcomes. A recent CEB survey of 10,000 employees at 1,000 organisations globally found that less than 50% of employees believe corporate information helps them do their jobs. Despite the hype, the rapid rise in data sources and potential uses may actually make it harder to capture real business value.
The problems are not technical. Big Data technologies are reaching the stage where, within limits, as much data as needed can be captured and integrated. However, our research found three other hurdles that organisations must be overcome before realising business outcomes from Big Data.
1. Find the business value needle in the Big Data haystack
Spotting the handful of genuine opportunities from the many dead-ends is hard. Too often, opportunities are identified from the bottom up based on the capabilities of a tool or the availability of a data source. The resulting initiatives are unfocused and unproductive. This is not to say that the opportunities aren’t there. During our research, we saw leaders in industries including automotive, pharmaceuticals, manufacturing, and banking, as well as in the public sector, using big data to cut costs, grow revenue, or increase customer satisfaction and loyalty.
For example, one automotive company combines data from its CRM, warranty and dealer systems, with data from Facebook, Twitter, YouTube and other social media platforms. By merging internal and external data, the company gets a richer and more timely understanding of potential customers, and can launch marketing campaigns faster, target messages more precisely, and see higher sales.
What’s interesting about this story, aside from the results, is how the idea came about. The company’s IT team sees cross-cutting data needs and data sources better than anyone else. While marketing wanted to exploit data from Facebook and other external sources, IT had a better understanding of data sources elsewhere in the company that could be included as well.
One of the company’s IT executives told us, “the value lies in tying social media data back to corporate data and placing it in context. Social media is one of many sources out there. Marketing misses a lot of the internal sources that IT is aware of.”
2. Make Big Data easy to use
The diversity of new data sources makes it harder to access data or present data in ways that improve decision making. Another reason why usability is a growing problem is that most information management initiatives are narrowly focused. In many organisations, information management means managing structured data in core systems, but only 50% of employees us this data, whereas more than 80% use data from sources such as data stored directly on desktops, data in the cloud, and unstructured data generated through knowledge work and in social media networks.
To extract value, IT must provide transparent and flexible access to information from diverse sources and design interfaces that let employees easily access, visualise, navigate, and analyse that information. Leading companies give their employees simple analytic tool interfaces that behave like familiar platforms such as a search engine, and present information in the wider context against which it can be understood. For example, a European telecom company improves data accessibility and usability through wild card searches, better presentation capabilities, and information contextualisation and enrichment.
Only 27% of employees find that prepackaged data and reports are valuable. The rest bemoan the time they waste reformatting and integrating reports before they can use them, and want to self-serve instead. This isn’t just a question of providing more tools. To self-serve effectively, employees need to know where the data is, how to get it, and how to use it. A global pharmaceutical company helps employees understand and visualize the interrelationships between data elements and provides full transparency into information sources, flows, and quality.
3. Teach employees how to exploit it
Only 38% of employees have the skills and judgment to use data effectively for decision-making. This may be the single most serious hurdle to capturing value from big data, but it is often overlooked by leadership teams.
This problem can only be solved through collaboration between IT, HR, finance, marketing and other information-intensive functions, and business unit leadership. IT can provide coaching and ensure that when new analytic tools or data sources are rolled out, employees know how to use the data, as well as the tool.
The information management team at a leading US retailer has embraced this idea. They hold regular roadshows to teach employees how to use new data sources. The emphasis is squarely on making better business decisions, not on learning the finer points of the latest analytic tool. They also recognise that data experts in IT should be able to teach employees, so they look for coaching skills when hiring for these positions.
Collectively, these three steps help to make business outcomes from big data a reality. IT has a role to play in all three and by doing so, ensures that the “I” in IT becomes at least as important as the “T”.