StoreRunner enables consumers to search millions of branded products across 6,000 categories including more than 700 of the nation's leading retailers. StoreRunner combines a variety of brands and merchants with the convenience of shopping online or at local retail outlets. Founded in 1998, the company's goal is to create the most comprehensive and useful shopping service ­ on or off the Internet.

Implementation

As a Web-based shopping portal that generates income from consumers clicking on ads and buying products, StoreRunner needed a system to accurately record site traffic and the huge volume of transactions that support its billing processes. While business intelligence reporting package systems such as Net Analysis could provide generic site information such as page views, visits and duration of visit, StoreRunner required more specific functionality. The company identified the need for a custom tracking solution to accurately collect clickstream information and customer-activity information.

With a significant portion of company revenue coming from customer activity on the Web site, executives realized the importance of recording every hit and transaction. In addition to recording the clickstreams of customers, there was a need to capture information about how each visitor accessed the site. Such information is valuable to the company's traffic-partner program, where StoreRunner.com shares clickthrough revenues with partnering companies that direct traffic to its site.

A team of consultants was hired in early September 1999 to build the solution. StoreRunner selected Informatica PowerMart 4.6 data integration solution for its automated extraction capabilities as well as its compatibility with Microsoft SQL Server and various other Web servers.

The biggest issue StoreRunner has faced thus far is staying focused on the end goal. Though the warehouse contains a comprehensive wealth of site activity data, it is vital to continually ensure the information is accessible to end users. For the system to be effective, it is imperative that customer clickstream data be made available to decision-makers and executives in a readable, reportable format.

One significant risk in building a custom logging solution is maintaining flexibility in the system. For example, StoreRunner.com launched a new user interface for the site on May 1, 2000, resulting in a new set of fields that needed to be recorded in its data warehouse. With an inflexible business intelligence system, tremendous development time would have been required to incorporate those new fields. By designing a warehouse that is event driven, the only necessary change was identifying which new events to track. Due to this level of flexibility, StoreRunner was able to adjust the system for the new site display in only a three-week time frame.

In terms of StoreRunner's traffic-partner program, a recurring challenge is in click-count reconciliation. Although StoreRunner's 1,000 partners use a variety of clickstream tracking applications, StoreRunner.com's internal numbers are invariably different from partners' figures. While the only solution to this issue is investigating the discrepancies on a case-by-case basis, an enormous timesaving factor has been the flexibility of StoreRunner's internally generated figures. The reports generated by StoreRunner effectively cut required investigation times in half.

StoreRunner initially selected Informatica PowerMart 4.6 to populate its data warehouse with clickstream information and later upgraded to PowerCenter 1.6. PowerCenter collects all transactional data that occurs on StoreRunner's Web site, including clickstream logs, banner advertisements and merchandise information. The company selected Informatica for its versatility as it is compatible with flat file and Microsoft SQL Server sources as well as the target Oracle database. With company-wide site activity information in a central data warehouse, executives and billing accountants use Brio to generate accurate invoicing and payment reports.

Tips

Recommended strategies based on StoreRunner's experience:

  • Build a flexible system architecture. With Informatica as its data extraction solution, StoreRunner was able to quickly adjust to changes in its site content.
  • Select BI technology partners. Avoid writing SQL or PERL scripts as they will likely lead to a nonscalable, inflexible system.
  • Focus on the practicality of the project. When developing a company-wide tracking system, it is easy to get lost in the technological components. Communicate with end users to provide exactly what they need.

Results

  • Accurate billing of merchants due to detailed site activity information.
  • Accurate analysis of visitor origins for traffic-partner program.
  • Ability to rapidly adapt system to changes in Web site content.
  • Increased knowledge of customer preferences and behaviors for marketing strategy development.

Informatica PowerCenter 1.6 - A data integration hub that enables large organizations to easily transform Web, legacy, relational and ERP data into reliable information for strategic business analysis.

Brio.Enterprise 6.0 - An application that delivers relevant information in real time to various users, from front-line managers who need an instant view of the business to external partners who need reports via their browser.

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access