As big data grows within financial services, two primary types of data are taking center stage: on-grid and off-grid.
By definition, on-grid is all the data a bank has about its customers — all the insight it receives from them when they talk to the bank, whether it’s in social media, web interactions or any other direct communication. However, off-grid data (representing 80 percent) is generated when customers talk about the bank without actually talking to the bank—for instance, when a customer relates opinions on Reddit or in a chat room.
Historically, banks have analyzed on-grid data only – but, in doing so, they need to understand how bias influences the data.After all, if a customer is truly unhappy, that person’s comments will be far more authentic in informal settings and when unsolicited or through informal channels.
The good news is that future opportunities are virtually limitless, especially in efforts to improve fraud detection and prevention—a huge priority in the financial sector. For instance, consider the possible creation of a utility that aggregates information for analyzing the entire industry for trends and patterns in fraud. When a new attack or scam surfaces, the information will immediately be shared with every bank. That way we would no longer see three banks discovering the same problem, respectively, months apart and having to deploy individual solutions. Just the cost and time benefits of independent discovery would make creating such a utility well worth the effort. This applies an approach similar to that of antivirus software in computers – once identified, all relevant points are informed and can act to prevent damage.
Similar benefits exist on an individual bank basis. For instance, analyzing data from a customer panel takes months, but big data analysis can provide far more insightful results in mere hours. Using social media as an example, mass comments can be analyzed in a far more timely and relevant manner, using views expressed a few minutes ago to offer a far richer data set.
As with many of these next-generation technologies, the time to value and realization of a proof point are crucial. Success starts with a limited investment, answering a very specific question and doing it well. It’s about demonstrating the outcome and what’s possible. From there, organizations can build upon their successes by replicating efforts in other areas of need. Fortunately, proof points are relatively inexpensive, and can be used in conjunction with more traditional (and more commonly accepted) methods.
Of course, big data has its challenges. The issue of leveraging the cloud to host or analyze data is a prime example. Financial institutions, in particular, are nervous about bringing data to the cloud as regulators continue to consider the impact of this not-so-new delivery model. And banks understandably prefer to err on the side of caution to avoid breaking future rules—not to mention putting their most valuable data at risk. The second challenge is that big data actually requires data. Whether through social listening, leveraging legacy data or collecting from industry-specific sources, banks’ ability to locate and extract the data into usable environments is key. Fortunately, unlike other industries, banks have far fewer challenges with the analytics component.
Banks are making progress, including a recent uptick in the number of chief data officers. This is an important step, because it provides governance and establishes ownership. However, big data remains an untapped opportunity. Even though great examples exist, big data use within financial services isn’t mainstream yet.
Have your say in the post comments section below
By definition, on-grid is all the data a bank has about its customers — all the insight it receives from them when they talk to the bank, whether it’s in social media, web interactions or any other direct communication. However, off-grid data (representing 80 percent) is generated when customers talk about the bank without actually talking to the bank—for instance, when a customer relates opinions on Reddit or in a chat room.
Historically, banks have analyzed on-grid data only – but, in doing so, they need to understand how bias influences the data.After all, if a customer is truly unhappy, that person’s comments will be far more authentic in informal settings and when unsolicited or through informal channels.
The good news is that future opportunities are virtually limitless, especially in efforts to improve fraud detection and prevention—a huge priority in the financial sector. For instance, consider the possible creation of a utility that aggregates information for analyzing the entire industry for trends and patterns in fraud. When a new attack or scam surfaces, the information will immediately be shared with every bank. That way we would no longer see three banks discovering the same problem, respectively, months apart and having to deploy individual solutions. Just the cost and time benefits of independent discovery would make creating such a utility well worth the effort. This applies an approach similar to that of antivirus software in computers – once identified, all relevant points are informed and can act to prevent damage.
Similar benefits exist on an individual bank basis. For instance, analyzing data from a customer panel takes months, but big data analysis can provide far more insightful results in mere hours. Using social media as an example, mass comments can be analyzed in a far more timely and relevant manner, using views expressed a few minutes ago to offer a far richer data set.
As with many of these next-generation technologies, the time to value and realization of a proof point are crucial. Success starts with a limited investment, answering a very specific question and doing it well. It’s about demonstrating the outcome and what’s possible. From there, organizations can build upon their successes by replicating efforts in other areas of need. Fortunately, proof points are relatively inexpensive, and can be used in conjunction with more traditional (and more commonly accepted) methods.
Of course, big data has its challenges. The issue of leveraging the cloud to host or analyze data is a prime example. Financial institutions, in particular, are nervous about bringing data to the cloud as regulators continue to consider the impact of this not-so-new delivery model. And banks understandably prefer to err on the side of caution to avoid breaking future rules—not to mention putting their most valuable data at risk. The second challenge is that big data actually requires data. Whether through social listening, leveraging legacy data or collecting from industry-specific sources, banks’ ability to locate and extract the data into usable environments is key. Fortunately, unlike other industries, banks have far fewer challenges with the analytics component.
Banks are making progress, including a recent uptick in the number of chief data officers. This is an important step, because it provides governance and establishes ownership. However, big data remains an untapped opportunity. Even though great examples exist, big data use within financial services isn’t mainstream yet.
No comments:
Post a Comment