Use Big Data to disrupt the world in your favour.
In 2013 IBM came out with the startling statistic that “90% of the data in the world today has been created in the last two years alone”. Since that time, the drivers of “Big Data” – credit cards, mobile phone calls, social media, photographs, streaming video etc. – continue to create a torrent of data and the metadata that supports it, and at an accelerating rate.
The growth of Big Data has been so fast that marketers do not really know how to harness it. “Getting information off the Internet (as a large and growing source of today’s data) is (truly) like drinking from a fire hydrant”.
The following post present some thoughts on how to take advantage of the advances in Big Data.
1.) Recognize its importance
While marketers have traditionally employed a mix of judgment and science (tracking studies, focus groups, planning) to aply their craft, Big Data poses both a threat and an opportunity. Those who become adept at using Big Data it will benefit at the expense of those who don’t.
The reason for these advantages is there’s a clear path that leads from data to analytics to visualization to insights, intelligence and ideas. This chain starts with Big Data leads to ideas and ultimately results in a change in consumer attitudes and/or behaviour.
2.) Learn the basics
Learning about Big Data requires a curious mind and actively engaging specialists on the subject - either internal to your organization, or external to the firm.
A good place to find experts on Big Data is to study companies that are excelling with Big Data. These can be found through desk research and working with your IT colleagues. In my most recent search on the topic Google points me to Fast Company’s “The World’s 10 Most Innovative Companies in Big Data”. This is a good starting point.
Another useful tool is to consider Bernard Marr’s 5 Vs of Big Data: Volume, Velocity and Variety, Veracity and Value. (Bernard seems to have added “Veracity” and “Value” to an existing list.)
This refers to the incredible amounts of data generated each second from cell phones, credit cards, cars, M2M sensors, photographs, social media, video, etc.. This data also needs to be stored and there is no one data base that can manage this enormous and growing task; rather, data needs to be distributed over a number of systems and available for ongoing access and analysis. (Thankfully storage costs have been plummeting in recent years so this is now economically possible.)
This refers to the speed that Big Data is collected and analyzed. An increasing number of “Always on” sensors from the office, factory, home and Internet of things, are producing a real and continuous torrent of data, 24x7x365.25.
While the growth of data is impressive, the speed of transmission and access must also remain instantaneous, to allow for real-time access to website, credit card verification and instant messaging. This is clearly an engineering challenge of vast and growing proportions.
This refers to the different types of data that are being collected, including text, sound, video, and the underlying meta-data including dates, times, location etc..
Some of this data, such as financial transactions, are stored in traditional data banks, but more often today's data, such as photographs, and video, are stored in unstructured databases and can be found through user applied tags, etc..
While there is an incredible variety to data there is also some commonality, because the vast majority of this data is digital and build from its component binary code.
This refers to the quality and/or trustworthiness of the data. While many think of Big Data having a high level of data, this is not always the case. For example the resolution of a photograph may be off, or the data from a machine may be contaminated with noise.
A good example of this concerns GPS data which “drifts” off course in urban settings because the satellite signals bounce off tall buildings. In cases like this location data must be fused with another data source e.g., road data or data from an accelerometer, to provide accurate data.
This refers to the value that you can extract from Big Data. While there is a clear link between data and insights, this does not always mean there is value in Big Data. Just because you can measure something does not mean that it should be measured.
A simple rule of thumb is to start with your Objectives, Goals, Strategies and Measures (OGSM) and see how these can be enhanced by Big Data. Then start to experiment.
A good place to experiment is to use Big Data to crack your most difficult problems. (In other words try to fix what’s broken and leave what’s working alone, at least for the time being.)
Another good place to start is to apply some of the successes that Big Data has to fixing other businesses, including:
a.)Challenging the experts through crowd-sourced data.
In many cases Big Data is proving to be more reliable at predicting results than experts in your field. For example, AC Nielsen provides rating guidelines for programs and companies such as Networked Insights mine the internet to find social sentiment which is often a better predictor of results than any rating bureau.
b.) Customer segmentation to reduce marketing waste.
Marketers can reduce the amount of Free Standing Inserts they waste by using Big Data to better target households. Recent technology with cell phones can be used to track back traffic flows to retail outlets, allowing marketers to better target those who visit their stores.
In the auto insurance business a number of carriers have found that correlations between credit scores and the speed at which an individual reacts to an issue e.g., Check engine light, is directly correlated to his risk. This data can help the auto insurance attract a better customer to its business and reduce its claims ratio – directly helping the bottom line.
As there is uncertainty in results, it is probably best to start with 2-3 experiments with Big Data vs. putting all your eggs in one basket.
4.) Build on your successes
Given that the investment and results related to Big Data can generally be directly measured one can determine the returns quite quickly. In turn you can build on successes and reduce investment in failures early.
While ignoring Big Data may not immediately kill your business, you can’t ignore it for long. While your competitors might not be using it, competition now comes from many sources. Big Data is one of those tools that is proving to be very disruptive.
It’s time to stop thinking of how long you can avoid this trend to identify what can be gained from this amazing phenomenon. Use Big Data to disrupt the market in your favour.
The Rising Sun