Instruction: Attempt all scenario based questions

ANSWER-

1.

Card information is stored in a number of formats. Card numbers – formally the Primary Account Number (PAN) – are often embossed or imprinted on the card, and a magnetic stripe on the back contains the data in machine-readable format. Fields can vary, but the most common include: Name of card holder; Card number; Expiration date; and Verification CVV code.

In Europe and Canada, most cards are equipped with an EMV chip which requires a 4 to 6 digit PIN to be entered into the merchant's terminal before payment will be authorized. However, a PIN isn't required for online transactions. In some European countries, if you don't have a card with a chip, you may be asked for photo-ID at the point of sale.

In some countries, a credit card holder can make a contactless payment for goods or services by tapping their card against a RFID or NFC reader without the need for a PIN or signature if the cost falls under a pre-determined limit. However, a stolen credit or debit card could be used for a number of smaller transaction prior to fraudulent activity being flagged.

Card issuers maintain several countermeasures, including software that can estimate the probability of fraud. For example, a large transaction occurring a great distance from the cardholder's home might seem suspicious. The merchant may be instructed to call the card issuer for verification or to decline the transaction, or even to hold the card and refuse to return it to the customer

1. Turn on gateway fraud filters.

Most payment gateways allow ecommerce merchants to set up some basic fraud prevention rules to block or flag transactions that may be fraudulent. Typical examples are to decline all transactions when the billing address does not match what the credit card company has on file (an AVS mismatch) or excluding all transactions from specified countries.

Pros: No cost to merchants.

Cons: The selection of available rules is inadequate for merchants facing a moderate to high amount of fraud. There is also a high risk of false declines as the rules are not very flexible.

This step is recommended for merchants in low-risk fraud categories and merchants that have not experienced much fraud.

2. Manual review.

Most ecommerce merchants have employees assigned to review purchases that fall into specific risk categories, such transactions above a specified dollar amount and/or transactions where the billing and shipping addresses don’t match. Common manual review techniques include using Google to find verifiable data on the purchaser, checking social media accounts and using Google maps to see if the shipping address appears legitimate. The average merchant reviews 26 percent of all transactions, of which 80 percent are eventually submitted for processing.

Pros: This is more accurate than simple Gateway filters, since a veteran employee becomes quite effective over time in preventing fraud.

Cons: Time consuming, results vary greatly according to employee skill, potential bottleneck during high season, risk of experienced employee leaving the company and limited tools available for research

This is recommended for merchants in low to moderate risk fraud categories.

3. Fraud prevention solutions.

There is a multitude of third-party fraud prevention companies (full disclosure: including my company, NoFraud) that leverage sophisticated fraud prevention technologies such as IP Proxy Piercing, Geolocation, Device ID and Global Fraud Blacklists to reduce or eliminate fraud liability. The solutions range from the very basic, providing a risk score and tools to build a fraud prevention algorithm, to full service solutions that leverage machine learning to give you a yes/no response and will even reimburse you for any fraud chargebacks that resulted from their decision.

Pros: Very effective against fighting fraud. Most solutions eliminate the need for manual review, providing expert service that is not tied to employee skill. Companies can accept more orders that would have normally been declined due to fraud concerns, allowing you to pre-determine fraud costs.

2.  Analysts say Billpoint failed because of a poor business plan, aggressive competition from PayPal, and hostility from eBay sellers. The acquisition was meant to help eBay speed the closure of auctions by allowing buyers to use credit cards to pay for their purchases.

In the early days of PayPal, its most important rival was Billpoint, a rival payment system that was a joint venture between eBay--PayPal's most important partner--and Wells Fargo Bank. Consider the situation PayPal faced: the vast majority of its business at the time consisted of handling payments for eBay auctions, yet eBay itself owned a competitive payments business (Billpoint) that it was promoting to every single eBay user. To outside observers, the circumstances must have looked grim.

Yet as we know, PayPal triumphed over Billpoint, leading eBay to purchase PayPal for over $1.5 billion. One of the key factors was PayPal's superior use of network intelligence. Reid led this intelligence-gathering effort for PayPal (he was executive vice president at the time) and asked all the members of the team, from executives to individual engineers, to use their network intelligence to learn about Billpoint's strategy. Billpoint's team, on the other hand, completely ignored the potential for network intelligence to provide insights into PayPal's strategy.

3.  

Amazon.com sells lots and lots of stuff. The direct Amazon-to-buyer sales approach is really no different from what happens at most other large, online retailers except for its range of products. You can find beauty supplies, clothing, jewelry, gourmet food, sporting goods, pet supplies, books, CDs, DVDs, computers, furniture, toys, garden supplies, bedding and almost anything else you might want to buy. What makes Amazon a giant is in the details. Besides its tremendous product range, Amazon makes every possible attempt to customize the buyer experience.

When you arrive at the homepage, you'll find not only special offers and featured products, but if you've been to Amazon.com before, you'll also find some recommendations just for you. Amazon knows you by name and tries to be your personal shopper.

Amazon Technology-

The massive technology core that keeps Amazon running is entirely Linux-based. As of 2005, Amazon has the world's three largest Linux databases, with a total capacity of 7.8 terabytes (TB), 18.5 TB and 24.7 TB respectively [ref]. The central Amazon data warehouse is made up of 28 Hewlett Packard servers, with four CPUs per node, running Oracle 9i database software.

Amazon E-commerce-

Amazon.com has always sold goods out of its own warehouses. It started as a bookseller, pure and simple, and over the last decade has branched out into additional product areas and the third-party sales that now represent a good chunk of its revenue (some estimates put it at 25 percent).

Amazon Tools, Marketing and Community-

The goal is pretty straightforward: "To be Earth's most customer-centric company where people can find and discover anything they want to buy online." The implementation is complex, massive and dynamic. Amazon's marketing structure is a lesson in cost-efficiency and brilliant self-promotion. Amazon's associates link to Amazon products in order to add value to their own Web sites, sending people to Amazon to make their purchases. It costs Amazon practically nothing. Some associates create mini-Amazons -- satellite sites that do new things with Amazon data and send people to the mothership when they're ready to buy. Amazon Light, built and maintained by software developer Alan Taylor, is one of those satellite sites.

I appreciate use of item-to-item collaborative filtering by amazon rather than traditional collaborative filtering. Here are some tips to improve recommendation and the hit rate of recommended product.

1. Combine the interests of the entire family: Some time (Like buying a sofa) is a combined interest of family but not just the individual. So it is better to combine interest of the family members for such items which indeed improves the recommendation. Hence ask new member to enter family detail(hence combine amazon account of family members).However, I am aware that there exists million users already. SO by looking at the delivery address of the users, one can group users as a family. This one definitely improve the recommendation engine.

2. Combine interests of friends: While buying, a person is likely to get suggestions about his selection from his close friend. I do consult my best friend before buying a jeans pant or a tank top. Hence by providing a share with a friend option, we can easily track down, whose opinions a person is likely to care. And then by connecting those profiles as friends and then combining their interest before making suggestion can improve the recommendation.

3. Color: Consider this variable seriously. Few people are likely to purchase an item just because its their favorite color and few get attracted to items displayed in their favorite color than the same item in different color. Hence, for any recommended product make sure that you Display the product in user's favorite color or very similar one if its not available. Favorite color of user can be determined by asking the user to enter while creating account or by learning from their previously purchased item's colors. this helps in increasing the hit rate of recommended product.

4. Priceline's other two revenue sources are agency revenues and advertising revenues. Agency revenues represent the fees Priceline charges service providers through the traditional price-disclosed model, when someone chooses one of their products offered on one of Priceline's web platforms.

The business model of Zappos is built on providing an exceptional customer experience, and in doing so, building customer loyalty and driving word-of-mouth marketing. ... Free shipping and returns on all orders, regardless of order size, enabling customers to order multiple items to try on in the comfort of their own home