5.7 Commerce Server Data Warehouse and Business Desk Analytics

 < Day Day Up > 



One thing you won't be short of on your Web site, unless you have no visitors, is data. Commerce Server can collect no end of data from the moment users visit your site to the moment they move on. The Commerce Server data warehouse can import, transform, and manipulate the data, and then have these data presented to business users via the business desk.

The data warehouse can implement the following activites:

  • Import Web log files, user profiles, transactions, campaigns, and catalog data from a Web site (Since the data warehouse is based on SQL Server, these tasks are completed using the data transformation services (DTS) tool.)

  • Prepare the data for analysis by populating the data warehouse (OLAP) cubes

  • Resolve Internet Protocol (IP) addresses to convert the addresses to readable domain names

  • Delete imported log file, test, or archived data to reduce the amount of storage space required within the SQL Server database

The business desk contains a number of modules that allow a business user to analyze site activity. There are a range of reports available out of the box, but others can quite easily be created.

Dynamic reports are those that are generated at run time, so that the most recent data in the data warehouse are used to populate the report. The data that are displayed are not actually stored. Static reports are run immediately, but the data is stored so that the results can be viewed later without rerunning the report.

5.7.1 Segmentation

This is a neat technology that allows analysis of a group of users having similar profile properties and site behaviors. Typically, they would have similar attributes, such as gender, age, or time they visit the site. Once a group of users has been identified, they can be actively targeted with some very focused offers or marketing activities.

5.7.2 Prediction

Once data have been collected in the data warehouse, these data can form the basis of a prediction model. This enables you to see which customers are buying which products and then start to predict trends and possible patterns in future customer activity.

The basis of this is the prediction model, which uses data that have already established a pattern to number crunch raw data using some very complex and peculiar algorithms. The end result is a prediction assessment of a certain behavior from customers analyzed by the model.

A useful implementation of this technology is to suggest to purchasers of a particular product that they may wish to purchase another-a typical cross-sell activity. A good example is a customer buying a gasoline lawnmower from your site that suggesting to this customer might also purchase some engine lubricating oil.

5.7.3 Payments and Commerce Server

Commerce Server does not ship with a mechanism to check credit cards or authorize payments for goods. Instead, it has a default payment component that can act as a temporary holding place for credit-card details so that applications can be tested. If you wish to fully enable a payment mechanism, you need to purchase a third-party add-in solution. But before doing that, it might be a good idea to understand the typical Web site payment processes available to you (see Figure 5.8).

click to expand
Figure 5.8: Typical Commerce Server payment architecture.

  • Internet real-time payment allows a user to submit credit-card details, which are then transmitted to a payment service provider, who, in turn, passes the details on to the bank holding the appropriate merchant account . The bank will check the credit limit on the card and that the billing address matches the one in the bank's records. The advantages of this system are that it is real-time, straightforward to implement, and relatively low-cost. The disadvantages are that a lost connection to the bank will result in lost sales since cards cannot be authorized, and the system can slow down dramatically at times of peak demand.

  • Real-time with batched payment in-house systems are deployed internally to process payments, removing the need for the customer to wait for authorization . The credit-card details are batched together and then sent across to a bank in one go. The customer receives a message saying his or her order is being processed, and if there is a later authorization problem, the customer is contacted via e-mail (see Figure 5.9).

    click to expand
    Figure 5.9: Batch payments offer a low-cost entry point for a Commerce Server Web site.

  • Redirection services are used by some organizations . As the user pays for the products, her or she is redirected to another site that captures the payment information. The level of payment transparency available to the user does vary, and in many cases the customer is redirected wholly to another Web site, leading to a rather poor experience. The benefit is a much lower transaction cost since these redirection sites can benefit from scale of use. Security is also of concern since site spoofing and URL manipulation can lead to a host of problems generally not present in fully contained sites. In practice, Commerce Server is not geared up to use this type of service without additional coding and development work.

Ultimately, the final choice will depend on customer requirements, but remember that Commerce Server stops at the default payment component, and everything else will need to be written by the developers.



 < Day Day Up > 



Microsoft  .NET. Jumpstart for Systems Administrators and Developers
Microsoft .NET: Jumpstart for Systems Administrators and Developers (Communications (Digital Press))
ISBN: 1555582850
EAN: 2147483647
Year: 2003
Pages: 136
Authors: Nigel Stanley

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net