Planning for the Business Analytics System
You use the Commerce Server 2002 Data Warehouse to collect day-to-day operational data about users who visit your site, including user profile data, transaction data, and click-history data. You use the Analysis modules to analyze the data. For example, you can identify user trends or analyze the effectiveness of a campaign, and then update your site to target specific user groups, sell specific products, and so forth.
The following table lists some of the questions that you need to answer in planning for the Business Analytics System.
|How will data analysis affect the key business decisions you need to make about your site? What reports do you need in order to make those decisions?||To examine the reports available from the Business Analytics System, see Commerce Server 2002 Reports. If the reports do not meet your requirements, see Creating Custom Reports to learn more about modifying existing reports or creating new reports. There are also a number of third-party software vendors who provide reporting solutions that interact with the Business Analytics System.|
|Who in your organization needs access to the data? How often do they need to run the reports, and how much detail do they require from updates?||You should establish a schedule for running reports. You may need to run static reports on a regular basis with your Data Transformation Services (DTS) tasks, based on anticipated business requirements.|
|How often will you process your Web server logs?||The size and nature of the data stored in your logs will determine your answer. The more frequently you process the logs, the smaller they will be.
Larger log files provide a richer data set, but they also take longer to process.
There is an associated overhead with each post-processing task. In larger sites, it is recommended that you perform fewer log import and post-import processes each day due to the post-import processing that occurs. The log import process takes longer as the size of your Data Warehouse increases.
You should determine the appropriate processing frequency in your test environment, and then apply it to your production environment.