The difference between data and information: extracting real value from big data
Every second an enormous amount of data from around the globe is amassed. Companies need professionals with expertise in processing and converting this data into knowledge that will feed their decision-making and define their business strategies.
Organizations are increasingly basing their decisions on insights they gain from analyzing large quantities of data. Last year in Spain, the demand for big data professionals reached 6.413, representing a 17 percent increase from 2017, according to the Infojobs annual report on the Spanish labor market. Data is essential in order to define effective policies and strategies, but it doesn’t add value on its own. It needs someone to process and analyze it in order to convert it into useful information for businesses, government agencies, and other organizations.
Rubén Casado, senior manager at Accenture and director of the big data master’s program at Kschool, uses an analogy to explain an essential difference, “In a kitchen, data would be the ingredients and information would be the main dish.” The same data can deliver different pieces of information: “Data provides intelligence, which is added to this jambalaya and served to answer specific questions.” The big challenge for companies, therefore, is to give data meaning, transforming it so that it really and faithfully tells a useful story about the business, its customers, or society in general.
Each second, millions of data points are collected. Thousands of sensors installed across cities around the world collect data about all sorts of topics: air quality, noise, the temperature, even about which parking spots on different streets are free or when specific gardens need watering. Users also generate data: everything they do on their smartphones, every interaction they have with any business is tracked and stored. Any text or image published online can also be processed as data in order to extract information.
Work undertaken by BBVA Research to assess the perception of the fintech phenomenon provides a good example of the latter. BBVA Research put big data and artificial intelligence techniques to use in order to analyze information published on Twitter and other media outlets. “Thanks to technological progress we have all sorts of data gadgets and tools at our disposal. Translating this into knowledge is the role of the analyst, who needs to make sense of a morass of unstructured data and process and analyze it so that it becomes knowledge and drives growth and social well-being,” explains Tomasa Rodrigo, head of big data at BBVA Research and one of the supervisors of this project.
Data can be divided, depending on type, into structured and unstructured. The former is that data which comes with a structure already in place. “For example, a person has a first and last name and a date of birth, and all the data I receive about people will always come with this data. This is easy to process because you always know what you’re going to find,” Accenture’s expert adds. Conversely, there are unstructured formats like text, audio, video, and photography. Casado points out that before analyzing this kind of data, it is impossible to know what it might contain.
“Translating this into knowledge is the role of the analyst, who needs to make sense of a morass of unstructured data and process and analyze it so that it becomes knowledge and drives growth and social well-being,”
To extract information required for a business, internal data — the company’s proprietary data — or external data can be used. For example, if a business wants to see if public perception about them is positive or negative, it can process external data from Twitter or other social media platforms. “Or, to optimize when and where snowmobiles should be best allocated, weather predictions should be gathered,” Casado explains.
In a business context, big data and data science are used to analyze unstructured data that leads to a better understanding of the environment in which a business grows. This understanding is made more complete by analyzing social dynamics or geopolitical trends and how they interconnect with the economy “using data from social networks and the media,” BBVA Research’s Rodrigo adds. Another example of this kind of work is a BBVA Research study that used machine learning techniques to analyze, monitor, and measure central bank online publications including reports, press releases, meeting minutes, and other communications. This analysis provided greater understanding into central bank communication strategies.
Companies collect data because they want to get something out of it. For example, they want to predict how many customers they will have next month. “To get this information, which has commercial value, historical data has to be available: all past purchases, who made them, on what dates, what products were in the catalog at the time,” explains Casado.
All this data needs to be analyzed in order to gain insights that enable better decision-making. “Big data is a series of technologies conceived to enable very fast, real-time processing of large quantities of data,” Casado continues. There are various tools that enable different ways to access and store data, to process it, and exploit it.
In the banking sector, there are several projects based on big data. BBVA is working to extract value from data through initiatives like the AI Factory. The AI factory is the global development center where the bank grows its artificial intelligence capabilities, both for customer-facing solutions as well as for the optimization of the Group’s internal processes.
The bank has developed platforms that leverage data to improve the customer experience by providing personalized products. For example, BBVA Valora is a service that provides the approximate sale price of a new home, checks prices of comparable homes in the area, and lets customers know the potential impact of the purchase on their personal finances. BBVA is also working on developing increasingly customized data-driven solutions, which allow the bank to give customers tailored guidance about savings or advice about which financial products are most appropriate for each person. “Data doesn’t merely allow us to recommend the appropriate product for each customer, we can also give personalized guidance. For example, how can we help if customers’ finances aren’t getting them through to the end of the month? With Bconomy, using customer transactional history, we can already today warn customers when they are going to be short for their anticipated expenses at the end of the month. And more and more we can provide them practical advice, such as to transfer money from one account to another or to request an advance,” explains Gonzalo Rodríguez, global head of Customer Solutions at BBVA.
“Data doesn’t merely allow us to recommend the appropriate product for each customer, we can also give personalized guidance”
Concurrently, big data technologies are making it possible for banks to process all its transactions in real time. According to our expert, this is one of the pending challenges facing several industries. “Traditionally, everything was done one day for the next. But with each passing day, we are able to do more in real time. It would be normal to process everything in real time because that’s how things happen, they don’t happen in artificial six-hour or one-hour windows,” Casado asserts.
Other interesting stories