You can hardly attend any IT meeting or event without hearing the term “Big Data”. This term would have us think that it is Indian Jones’s King Solomon lost Goldmine and those that find it and unlock its secrets would be rewarded with riches beyond their dreams.
But in reality, how true is this ? Does data-mining really hold the secret-sauce to the company’s future ?
Futurist tell us that the ideal job of the future will be in being able to understand and explore Big Data to mine out the nuggets of information. I had the opportunity to meet with the KPMG Data Squad (not their official title but it should be) to discuss what KPMG is doing about Big Data. The Data Squad consisted of Frank Rizzo, data and analytics leader for KPMG South Africa, Karin Kruger, Associate Director Risk Consulting, Ntsiki Mpulo, Communications Manager.
Big Data is Old News
Data collection is not new. Neither is analysing data. Companies have been collecting and analysing their internal data forever in the form of sales and marketing reports, industry analysis, market segmentation, competitor analysis etc. What is new though is the source of data.
In the past the source was internal information, however now information is being delivered from external sources such as Internet, Social Networking, Data Hubs, Data Aggregators. Technology has allowed for systems to automatically interrogate data sources such as city data or customer social fees and even transcribed voice calls. All these data sources are now available at the finger tips to create better overall customer profiles.
Garbage In Garbage Out
Frank Rizzo says that “The three Vs of data examine its volume, variety, and velocity. At KPMG, we also talk about two additional Vs; the veracity (or integrity) of data, and the value that it provides the business. All these elements need to be used in conjunction to help decision-makers respond to a specific business problem they need to solve.”
The old GIGO (Garbage In Garbage Out) terminology still applies. The quality of data is paramount as analysing flawed data will results in flawed outcomes and therefore ensuring the integrity of the data has become an industry by itself. Frank explains that “It seems that we are all trying to solve the world’s problems with data and not many are checking the quality of the sources. Adding to this challenge is the amount of real-time data being generated. So the variety and volume of data have increased massively.”
Companies are now better places to understand a segment of the market on a macro level and down to individuals on a micro level.
So big deal – now what ?
Just because you can get data doesn’t mean you are empowered. One need to be careful about getting lost in the data. There is so much information that companies can spend time and money exploring data for data’s sake. The team at KPMG suggest that business need to drive the data by asking the business related questions. Starting with the end in mind – what are we looking to achieve? Then, when the questions are defined, let the data scientist loose of the data to find the relationships between various data sets.
KPMG places emphasis on the word AND that comes between data AND analysis. You need to collect the data but then you need to analyse it. These are two different tasks which might not necessarily be done by the same people at the same time.
Being able to extract information can lead to new opportunities as Safaricom discovered when they realised that their equipment on top of their cell phone towers was a better weather predictor than the weather service. The data the equipment was sending back was analysed and shared with the local farmers.
A cell phone company became the weather channel.
KPMG’s Dutch team was analysing the data patterns on social network and realised that there was a lot of movement to an isolated town. This was relayed to the police who investigated and discovered that someone had accidentally posted an open invitation to a house party. The same technology was then used during the Dutch Coronation where the police were able to monitor people’s location and sentiment so they could deploy officers to areas that demanded more attention.
The Data from Social networks used for real-time security.
Big Data is Big Business as companies are finding new opportunities in linking data sets together from quality verified sources of information. From being able to serve the right ads to the right people at the right time, to being able to predict patterns of where the next property boom will occur – this is where Big Data yields its rewards.
A South African company already in this space specialising in data analysis is Syenap. By using video technology they have the ability to track in real-time movement inside bricks-and-mortar environments that allows the retail owner to understand, predict and impact customer activity and enable strategy alignment. Having a better understanding of customer behaviour allows for better customer service. For example understanding how customer shops at a given display stand at a clothing retailer allows the retailer to predict when to restock the displays, thereby increasing sales (and not irritating the customer). Collecting the breadcrumb of every customer that walks in our view provides an encyclopaedia of data. However the real power is from analysing this unique data source as well as integrating with existing internal data sources. This Big Data analysis allows to predict future activity so that preventative actions can be taken before the event occurs. Knowing how many people will enter the establishment between 9 and 9:30 next week Tuesday allows for better staff planning to ensure adequate staff are there to man the till points and therefore reducing the time people spend in the queue.
Most importantly, the solution is fully compliant with all privacy legislation as customers are not individually identified but rather clusters of movements are analysed.
Adapting a movie saying I sum up the sate of Big Data as: “its better to have the Data and not need it then need the Data and not have it”.
We don’t know what we don’t know.
We don’t know what information we would need to be able to extract value out of the data and therefore companies are not discarding anything. Collection of data places additional burden on storage however these storage lakes are becoming cheaper as more are being used.
The scientist of the future will extract the value out of the data we keep today. Companies are starting to realise the necessity of analytics but quality data needs to be there first. A lot of experimentation is still happening in the South African marketplace around the real-time analysis of data but, according to Rizzo, there have been no massive successes as yet.
* headline image from Shutterstock.com