In 2011, U.S. telecom giant Verizon filed a patent to allow the company to watch viewers while they were watching TV through their set top boxes. As perturbing as this sounds, the idea, according to Bernard Marr, author of Big Data: Using Smart Big Data, Analytics and Metrics to Make Better Decisions and Improve Performance, was to analyze the conversations and offer improved target advertisements.
In the furor to create hyper-personalized customer experiences, it’s easy for companies to lose track of what kind of data is actually useful for their business or industry. As data keeps growing exponentially, access to terabytes and petabytes of data does not necessarily mean better insights and outcomes. To paraphrase what Marr says in his book, data is only useful if it answers the smart questions you need answers to. This in a nutshell is what ‘smart’ data is all about.
To understand the significance of smart data, we need to get a better perspective on the amount of data generated in the current environment. According to a report in Wired magazine, “Every minute, 48 hours of video are uploaded onto YouTube. 204 million e-mail messages are sent and 600 new websites generated. 600,000 pieces of content are shared on Facebook, and more than 100,000 tweets are sent.” Data from social media platforms is only the tip of the iceberg. Customers leave a slew of digital footprints when they search or shop online.
As companies amass a wealth of big and small data, they can conduct Exploratory Data Analysis (EDA) to eliminate the ‘noise’ and leverage ‘relevant’ data for better outcomes. To do this, it’s imperative to first discern the five aspects of smart data — volume, velocity, veracity, variety and value.
Take for instance the utility industry where a humungous volume of data is generated through smart meters, smart grids, and digital channels including social media and online transactions. EDA helps eliminate irrelevant data and distill the relevant data which can be used to build models to solve future challenges. Data from smart meters, for example, can help utilities identify the historical usage patterns for a customer and accordingly offer demand response management services. In such cases, the data from transactions can be excluded. However, the data from transactions can help segment customers who make payments on time, but miss a payment here or there.
Velocity is the speed at which the data is generated. This data can help in real-time monitoring and enable companies to send engineers to problem sites quickly using a set of rules based on historical data analysis. Following the real-time analysis, the relevant data can be relayed to other areas for deep analysis to develop predictive maintenance models. The insights drawn from these models can be re-routed to real-time monitoring systems. Such a process enables utilities to proactively manage maintenance, improve reliability, reduce unplanned service work and mitigate risks.
Variety refers to data from multiple sources such as marketing channels, social media networks, sensors and Customer Relationship Management systems. Some variables found in these data sources could be repetitive and may not serve certain functions of a utility company. But by leveraging analytics, these variables can be filtered.
Apart from being analyzed individually, data from various sources can be clubbed and utilized to generate insights. Data from sensors is one such example. A utility company that uses the data from sensors to check the pressure from gas lines can correlate it to customer service billing records to detect anomalies. This gives the company an edge to proactively curb theft.
Veracity means companies will have to parse the data to see whether it can be trusted. A customer who frequently makes late payments may take to Twitter to vent when the service is cut off. Does this mean the customer is right?
Value, as the name suggests, is the true worth of the data collected. It ensures that right insights are generated from the data and they lead to measurable improvements in business outcomes and customer experience. The selection of relevant data would be in direct correlation to the outcome the company is looking for to maximize value.
Hence, when a company is looking to fix a problem, all the five aspects of smart data need not be considered. By narrowing down the ‘right kind’ of data, businesses can apply the right metrics and methodologies to analyze this data, glean the more granular or subtle details, and get accurate results. It also makes the data less overwhelming and saves time.
With the combined power of advanced data analytics and visualization to explore data sets, companies can extract intelligent insights and detect new trends as well. Smart and fast are the buzz words around data that will help companies survive disruptive trends, and perhaps even begin new ones.
Data will always be critical to drive businesses, but it’s up to companies to be smart about asking the right questions and eliminating irrelevant data. Remember, in the invaluable words of Tim Berners-Lee, inventor of the World Wide Web, “Data is a precious thing and will last longer than the systems themselves.”
Know more about smart data — a joint effort by WNS DecisionPointTM and Knowledge at Wharton.