What banks can learn from Google and Facebook

Big Data is huge right now, but many companies are struggling to harness the potential of this new phenomenon. Chief among those slow to adapt to new methods of parsing information are banking institutions, which are in danger of being left behind in this brave new world of data management.

At the Banktech Summit that was recently held in Sydney, a discussion of this very issue acknowledged that the historical method of information-gathering by the banking industry has been through credit card data. This system has helped banks create products that would appeal to their customers, but it isn't the only game in town anymore. For the most part, though, banks haven't changed their habits.

Financial institutions aren't the only ones that grapple with the best way to manage and utilise vast amounts of data. A new study by data management giant Oracle notes that many companies are aware of Big Data's potential but are at a loss regarding the best way to use the mass quantities of data they are now gathering.

Tech firms are all over Big Data
There are a handful of exceptions to this situation. Google (Nasdaq: GOOG) and Amazon.com (Nasdaq: AMZN) stand out as companies that really know how to gather and utilise data. Google, for instance, was instrumental in the development of Hadoop, an open-source platform that excels at analysing data, as well as MapReduce, used in conjunction with Hadoop. And who can forget the fine Google paid for gathering too much information for Street View, not to mention the latest US$23 million assessment for tracking Safari users? Anyone who uses the giant's search engine knows, through Google Suggest, just how closely the tech firm is observing every keystroke users make.

Amazon also has access to oodles of information and uses Hadoop to collect raw data on one level, more structured information on another, and, finally, consumer behavioral information. The three levels allow for quicker sorting and implementation of the data than would otherwise be possible.

Social media is another way to collect information on customers, and it is here that Facebook (Nasdaq: FB) is unrivaled. The company is currently working with banks to develop social banking services on its site, and many businesses, including banks, have a substantial presence on Facebook. No doubt, having access to nearly 1 billion users and their profiles is the kind of environment for which most businesses would pay up.

In addition, Facebook's built-in data collection system, which can track companies' pages for user hits as well as the number of shared conversations about their brand, can be a great source of information. It's even possible that the type of monitoring Facebook is currently participating in to detect criminal activity could be used to collect marketing data as well.

What are banks waiting for?
Despite the myriad uses Big Data could provide for the banking industry, these institutions are somewhat sloggy in their implementation of the technology. Not only would using such information take the guesswork out of the creation of banking products for consumers, but it can also be used for risk reduction.

Other banks are also beginning to explore new possibilities in this area. Citigroup (NYSE: C) has been surveying users on its Facebook page on their feelings about social banking on the site, presumably in preparation for offering its services in conjunction with the social-media site sometime in the future. Commonwealth Bank (ASX: CBA) has teamed up with Facebook to develop a banking app for the social media site.

Banks have access to vast quantities of data created each time a customer completes a banking transaction or uses credit and debit cards. Since banks already have this information on site, they only need to figure out how to parse it and use it to create more revenue.

This latter is the part that stymies, but there are several companies that can do the analysis for them, including Google and Amazon. The biggest banks have the advantage, with large numbers of customer transactions to examine and model. Some companies will use Hadoop to discover changes in consumer behavior that will affect banks — such as the trend during the last recession for homeowners to stop paying their mortgage while keeping other debt, such as car loans, current. How might banks have reacted if they had known early on that, for the first time, people were more likely to pay their credit card bill than their mortgage payment?

Data management has shown itself to be useful to banks both for risk management and for new product development. They have the data, and the money to pay data servicers to crunch the numbers for them. All they need now, it seems, is the will to get on with it.