Topic 9

The industry that I have chosen has been banking....

We all know how credit cards work, but we never really see what is behind the scenes...

How does your credit card get approved and how does business intelligence play a role?

Lets use a scenario: Purchasing groceries at the grocery store

When you go to purchase the groceries, you are initiating a real-time POS transaction. (POS = Point of Sale, aka swiping your card)

The transaction gets routed to the issuing bank and run through a database to determine if this transaction is legit. If a website has had considerable fraud, it may decline the transaction right up front.

The unique part is next. Once the issuing bank agrees that the purchase is acceptable, it goes to a rules engine. This rules engine looks at a number of criteria.

This includes but is not limited to...

1) Is there enough available credit?
2) Is this a normal transaction for the customer?
3) Is this transaction in an appropriate geographical area?

Etc...

Each company has their own rules engine.. it is how that company detects fraud.

The technology that supports this process is fairly simple.. it is all about CONNECTION,

IN ORDER FOR THIS SERVICE TO WORK THEIR NEEDS TO BE A CONNECTION:

If there is not a connection.. the technology fails.. this is considered data unavailability...

In the event that the transaction is slow... it is called latency.. this can be caused by a poor connection..

BENEFIT: Simple fraud detection techniques can save the issuing bank and their customers millions upon millions in fraud a year.....


----------------------------------------------------------------------------------------------------------------------------------

The next business process is underwriting:

It is important to remember that technology is a tool and not an end all solution. Underwriting is just one example that can bever be 100% completed from technology. Computers can only think in black and white. They are only capable of making decisions based on the criteria that is programed. There are definitely shades of grey with underwriting.

The end decision can be determined from an individual who is able to think for themselves. There is absolutely no way to determine if a customer is going to successfully satisfy the loan. There are ways to gauge it by.

Here is an example of how it works...

1) Initial Application is taken...

This can been 100% online or a banker can input information into the computer..

Once the information is in the computer it is important to do some more homework...

2) Trust but verify...

The customer is not always going to give accurate information. They could be inflating their income or not stating liabilities. The computer system automatically requests documentation and runs it through a decision engine. This includes but is not limited to Tax Returns.. Verification of Employment.. Verification of Rent.. Appraisals... ETC...

3) Initial underwriting is completed.

The computer at this point has run the decision engine on its own and can recommend a loan for approval or flag it for manual review.

4) Approve, Approve with stipulations, Decline..

The underwriter can put final approval on the loan, request additional information, or decline the loan...

Some advantages of the system..

Keeps underwriters fresh. Not as much busy work to be done. Underwriters are assisted in finding potential problems that may arise.






Sources:

http://www.gravic.com/shadowbase/uses/realtimebusinessintelligence.html