Big Data in Banking

The Banking sector is a very good example of how Big Data has revolutionized the customer experience in Banking. There are days when we use to stand in long queues for depositing money to bank account, withdrawing money, sending money to some other account and so on. But in today's times we can do all such things from comfort of home , we don't even need to go out of our house for doing these things.We just need an average internet connection and within few clicks these operations can be done in just few seconds now.

This has given many opportunities to banks as well as customers to know each other in a better way . As online is the new normal now in the banking sector so the banks who haven't adapted this structure are on the verge of ending their Business and many have ended as well.

In online banking , operations happens in just few seconds (or we can say in real time) . That means data gets generated at a very fast rate. And we know if data gets generated fast then it needs to be processed fast so as to gain real time insights from the data and this is where Big data becomes an important part of Banking Domain because it is very easy to do real time analysis using Big data technologies . 

Major Applications of Big Data in Banking:-

1) Customer Segmentation:-

Using Big Data Banks are having complete information about there customers. This is not limited to just having  Customer's name,contact and address information. But today using Big Data Banks know very interesting things about there customers like:-

• Search pattern of each customer.
• Customer spending pattern.
• They know how happy a customer is from their services using feedbacks.
• They know what services customer is looking for and on what other services he might be interested in.

And so on . Based upon the above information banks divide their customer into segments which is also known as Customer Segmentation. It mainly happens on the spending patterns, habits, interest etc of various customers. After dividing them into segments Banks run targeted Campaigns for each set of Customers.

For e.g. :- If n number of customers are looking for loan. Then using Big Data Bank will recognise all those customers and put them in a particular segment (cluster) and run a targeted campaign for them which is about loan only.

And in today's time Big Data has emerged so much that it is even ending the Customer Segmentation.

Banks are now able to recognise each and every customer separately and they are able run separate campaign for each and every Customer based on his hobbies,interests,patterns etc.

2) Fraud Detection:-

Today Banks are recognising frauds in online transactions with help of  Big Data technologies. If we try to do this using old technologies it will take days to recognise a fraud but we have to do it in real time because it will hardly take 3-5 seconds for a transaction to complete And a transaction if completed can't be reversed. That means banks will have only 3-5 seconds to recognise that a transaction is fraud or not and in the same time they have to complete the transaction as well.

In actual, banks use a combination of OLTP and OLAP to achieve this goal. OLTP server is used for authorization and updation of  account balance and OLAP is used to store various transaction patterns which will recognise that a transaction is fraud or not .

Let us understand this application of Big data using an imaginary Use Case:-

Obit bank has started online banking services 5-6 years ago . But due to fraud transactions it faces a loss of 5% of its revenue every year. Dyna is a customer of Obit Bank. She often used to do most of her banking activities online like everyone else is doing in today's times. Dyna went to a retail shop and bought some items. After that she made a payment using her debit card. As soon as she swiped her card and gave her pin , internally a request has been sent to OLTP server . OLTP authorize the credentials and finds that credentials are correct. Now OLTP sends this request to OLAP server , it checks the transaction and compares it with various good and bad patterns stored in it. OLAP finds that it is matching with good patterns and sent an acknowledgment to OLTP server when OLTP server finds that everything is fine then it approves the transaction and balance got updated in Dyna's account . All this happens within 3-5 seconds. 

On some other Day a request got generated from Dyna's account for a transaction. The request went to OLTP server OLTP server authorize it and finds that everything is correct. It now went to OLAP server and it immediately finds  that it is not matching with good patterns in OLAP . So as an extra security check it made an automated phone call to Dyna to confirm that she has made that transaction or not . Dyna denies about that transaction and bank immediately cancels that transaction request. And like this way Obit bank caught a fraud transaction in real time and saves customer from fraud. 

3) Providing personalised services to every Customer:-

Lets understand this also with the help of above use case.

Dyna is a customer of Obit Bank. She wants to invest some money in Bank's mutual fund schemes . So she opens bank's app and start searching about it . She also wants to take loan for starting a new Business so she searched about loans as well. When she searched about all those things in Bank's App . Bank records all those activities of her and stored them and after few hours bank process the data from those activities and starts sending notification to her like:-

 "Hey Dyna, curious about starting a new Business, we have loan plans at very low EMI's. Just click here to check'em out"

she also got one more notification like:-

"Want to save your money for future start investing in our balanced equity funds, to know more click here" .

So you can see that how Bank used Dyna's search pattern to know her in a better way and starts providing personalised services to her.

So from above scenarios we can see how Big data has completely revolutionized the banking domain and in the upcoming time it will upgrade more and more.