A Question of Big Data
One of the current catch phrases or buzz words today is ‘Big Data’. Specifically, using technology to mine and analyse multitudes of data in a way that a human cannot and to gain new business insights that previously were not possible. Many companies are prepared to invest large sums of money to find hidden gems and nuggets of gold.
But can big data analysis achieve this? No, is the answer according to Danish business consultant Martin Lindstrom, a former brainchild at toy maker, Lego. According to Lindstrom, big data is useless unless it is complimented with small data (experience and intuition).
In order to mine big data, one must start with a hypothesis that can be researched and tested. In order for this to occur, the people with the experience, knowledge and intuition need to be formulating the hypothesis. However, what is actually happening is that the hypotheses are being proposed by the data miners and analysts themselves which is counter productive. The people who are providing (or at least hoping to provide) the answers are the same ones asking the questions and this recipe needs to change.
Essentially what this means to me is that undertaking a big data analysis project is still a good idea if there is a clear goal in mind, it is properly managed and that the business knows how to take advantage of any gold nuggets that are found. It should not be simply about fact finding.
Reference: “Think Small”, Aaron Watson, Acuity, July 2016, Vol 3, Issue 6.
One of the current catch phrases or buzz words today is ‘Big Data’. Specifically, using technology to mine and analyse multitudes of data in a way that a human cannot and to gain new business insights that previously were not possible. Many companies are prepared to invest large sums of money to find hidden gems and nuggets of gold.
But can big data analysis achieve this? No, is the answer according to Danish business consultant Martin Lindstrom, a former brainchild at toy maker, Lego. According to Lindstrom, big data is useless unless it is complimented with small data (experience and intuition).
In order to mine big data, one must start with a hypothesis that can be researched and tested. In order for this to occur, the people with the experience, knowledge and intuition need to be formulating the hypothesis. However, what is actually happening is that the hypotheses are being proposed by the data miners and analysts themselves which is counter productive. The people who are providing (or at least hoping to provide) the answers are the same ones asking the questions and this recipe needs to change.
Essentially what this means to me is that undertaking a big data analysis project is still a good idea if there is a clear goal in mind, it is properly managed and that the business knows how to take advantage of any gold nuggets that are found. It should not be simply about fact finding.
Reference: “Think Small”, Aaron Watson, Acuity, July 2016, Vol 3, Issue 6.