Use Multiple Data Sources for Full Customer Profiles
By merging the information gathered from their Web site traffic analysis tools with other data sources, such as customer databases, savvy online marketers can mine their Web site traffic data to maximize the effectiveness and the ROI of their Web sites.
The first step is to select a log file analysis software package that stores detailed Web site traffic data -- including individual visitor clickstream data -- in a relational database, such as Oracle or Microsoft SQL Server. The Web site traffic data can be merged with other data sources using a variety of business intelligence tools.
Next, you will need to select a vendor of business intelligence tools to provide the products and training necessary to merge your Web site data with your legacy data. Fortunately, the business intelligence software market is becoming increasingly competitive, so you should be able to find tools that meet your specific hardware requirements and budget.
When selecting a vendor, consider the amount of time you are willing to invest implementing the solution. If you have business intelligence software implemented at your company, using your current vendor makes sense. If you don't have a business intelligence vendor -- which is the case for many dot-com companies, you might want to consider vendors who distribute templates that allow you to create reports directly from the Web traffic data stored in the relational databases.
Once you have selected and implemented a Web site traffic analysis and business intelligence solution, what types of information can you mine from your Web site traffic?
Identify your Web site visitors by name, address and phone number, instead of by IP address or host name, through linking submitted form data to each of your visitors using cookie IDs. This information also can be linked with other legacy customer databases to provide valuable contact information for your sales team. For instance, by learning the identities of your Web site visitors, as well as their interests, targeted sales pitches can be tailored for each lead.
You also can link your legacy financial databases with your Web site traffic data, allowing you to determine how much money your Web site visitors are spending on your Web site, in your catalogs and through your toll-free number. With this information, you can measure which banner ads and Web links refer the customers that spend the most money.
If you found that one search engine referred visitors who spent twice as much as those from another search engine, for example, you would likely increase your advertising on this search engine. You also can compare the cost of advertising on different referring sites with the amount of revenue generated by each referrer to determine the advertising's effectiveness.
E-commerce Web sites commonly serve dynamically generated pages based on the interests of individual users. In addition, these sites often embed product numbers in the URLs. By stripping out these parameters from the URLs and linking them with your product databases, you can see which products generate the most interest, as well as which are being purchased. You also can e-mail your Web site visitors with special discount offers on products in which they expressed an interest, i.e., viewed product pages but did not purchase.
For an even more advanced analysis of your Web site traffic data, consider using an online analytical processing tool. Inexpensive OLAP tools can be used to slice and dice your Web traffic data in combination with other data sources to provide different views of the data and to drill up and down within the data. For example, while most Web site traffic analysis tools can identify unusual dips or peaks in your Web site traffic, OLAP tools allow you to determine the exact causes.
While e-commerce businesses such as online retailers have been the first to embrace these data mining techniques, it is clear that traditional businesses with an online presence can also reap the same benefits.