We use cookies and other similar technologies (Cookies) to enhance your experience and to provide you with relevant content and ads. By using our website, you are agreeing to the use of Cookies. You can change your settings at any time. Cookie Policy.

Retail Business Intelligence Extranets—Implementation Options

Originally published July 19, 2005

Last month, we talked about the business benefits retailers can realize through sharing detailed sales and inventory information with their supplier partners (Retail Business Intelligence Extranets Collaborating with Suppliers). This month, in the conclusion of the retail business intelligence extranet series, we will discuss the technological approaches a retailer might take to implement a business intelligence extranet.

There are three different approaches that a retailer can take to share information, each with its own advantages and disadvantages:

Option 1: Business intelligence extranet integrated with retailer’s data warehouse. In this model, the retailer literally allows suppliers to become users of its enterprise data warehouse. There are two different architectures that can be used within this model. The first is to physically share the internal infrastructure with external users. The second is to create a separate infrastructure from that used by internal users for security or performance reasons.

In a shared model, data must be secured such that suppliers cannot see other suppliers’ data and such that suppliers cannot see internal data that the retailer wishes to keep confidential. The advantages of this approach include one set of batch loading and updating processes, less physical infrastructure (number of servers) and less ambiguity about data sources and consistency/latency differences in multiple platforms.

This approach seems well suited to larger retailers with significant IT resources. Wal-Mart and Lowe’s, for example, take this approach with their suppliers, merely adding the suppliers as users of their already-substantial business intelligence applications. Data volumes are in the terabytes and users are in the thousands. These systems are high-end and expensive, but the businesses that deploy them stand to drive millions if not billions of dollars in savings and revenue enhancements, so the ROI is easy to justify.

In a separated model, it becomes easier to secure and compartmentalize data and the approach is perhaps more manageable from a performance standpoint during peak load times. Supplier queries can be prevented from slowing down the system for internal users. The disadvantages include support for multiple platforms (even if the technologies are the same) and increased development and maintenance of ETL routines to keep the separated databases synchronized.

A mid-tier retailer should consider this separated approach. There are still substantial cost and revenue improvements to be gained from setting up an extranet at this size of business, and the separated approach is easier to get off the ground because it can be taken on as a separate project from any internal data warehousing efforts. It is best when the merchandising and logistics departments drive the activity for this approach, especially in selecting a strategic subset of suppliers to be the initial users of the system.

In either architecture, the benefits include the ability to guide suppliers into the types of analysis a retailer wants them to perform, and the ability to allow a supplier and a merchant to look at the same physical report in the system to address sales or inventory disputes. The downside of the business intelligence extranet approach is that it is more costly than merely providing data downloads. This means more licensing for business intelligence tools—for this reason, the extranet approach is more appropriate for larger suppliers where small performance gains make a big difference because of the volume.

Option 2: Pushing reports or data to suppliers. In this model, the retailer pushes static reports or data sets to its suppliers on a regular basis. This is a bit of a compromise between creation of a complete business intelligence extranet and merely allowing data to be downloaded. The retailer can still guide its suppliers into the kinds of analysis it wishes to draw to their attention, but the cost of the approach is lower.

From a technical standpoint, a retailer could leverage the emailing capability that exists with many of the popular business intelligence tools to push finalized reports to its suppliers on a regular basis. The licensing costs for email recipients in business intelligence products are typically much lower than the costs for interactive or pull-based seats, and the retailer can tightly control the complexity of the requests to the database, as well as specify when they are executed to prevent overloading their peak usage windows.

A less expensive, but similar variation would involve creating data dump routines with some analysis built into them that FTP or otherwise deliver a flat file or XML output directly to a supplier. This approach could be as simple as SQL scripts and scheduled jobs, but it certainly puts the onus on the supplier to do something meaningful with the data. On the cost side, this variation probably eliminates the need for business intelligence tool licensing. If this is the only approach a retailer is considering, they are probably better off with Option 3.

Option 3: Provision of POS data downloads to suppliers. In this model, the retailer merely makes daily sales data files available to its suppliers for download. The supplier can load the data into their own systems for analysis and in some cases can import it into pre-packaged applications that accept POS data for analysis. This is the least-costly model for the retailer because the supplier pays for the processing cycles to load and report on data, business intelligence tool licenses to analyze it and the storage to build history for trend analysis.

The downside of this approach is that the retailer and supplier cannot view the same report, and the retailer has no control or influence over the kinds of analysis the supplier does to support their business. For these reasons, this approach is best taken with smaller vendors/suppliers where business improvements are probably small but important. Another benefit of this approach is the idea that the retailer is a better business partner to its suppliers, and can perhaps extract better terms in negotiations.

Some of the larger retailers are considering hybrids of all these options, and tailoring each “product” to the size of the supplier. Larger providers of goods get access to a dedicated application, and the smallest providers get data download capability. In all cases, both retailers and suppliers benefit from business intelligence extranets set up to give each a better view into their relationship. Real ROI is generated in the supply chain as a result, and improving technology coupled with declining processing costs are making the approach possible for a larger percentage of the retail community.

  • Dan RossDan Ross
    Dan is the Managing Partner of the Retail Practice at Claraview, a strategy and technology consultancy that helps leading companies and government agencies use business intelligence to achieve competitive advantage and operational excellence. Claraview clients realize measurable results: faster time to decision, improved information quality and greater strategic insight. Dan is a frequent contributor to business intelligence literature, writing on topics spanning technical approaches and business impact, and the Claraview Retail Practice serves some of the world's most advanced users of retail data warehouses.

    Editor's note: More retail articles, resources, news and events are available in the BeyeNETWORK's Retail Channel. Be sure to visit today!

Recent articles by Dan Ross



Want to post a comment? Login or become a member today!

Be the first to comment!