An Entrepreneurs Guide to eMarketing: Strategy and Implementation

 Version:  5 October 1999

Copyright © About Rob Kleine 1999, All Rights Reserved.

On-line Store Design Options: Making Product and Service Information Available on the Internet

This section provides a brief overview of three approaches to building a web store.  Emphasis is on describing the strengths and weakness of the three prototypical approaches to building web stores.  I shy away from tying the approaches to specific technology because the underlying technology is changing so fast and because the approaches described are intended as rough sketches and not detailed how to do it

Static Content Web Store

Static content web stores are similar to print advertisements and catalogs in that the content is published in a fixed form.  The biggest difference is that the web store happens to be published to the web rather than to paper.  Unlike ads and catalogues, web pages are available 24 hours a day to prospects worldwide.  As with catalogs and print ads, everyone who visits a web static store views the same information.  Also like print media, to change the content of a static content web store you must change the source document and then republish the catalog or advertisement.  The publishing process looks something like this:

The greatest strength of the static content approach is that it requires the least technical background.  Almost anyone paired with one of the many excellent web design software packages now available, could have a store on line in an afternoon.  Heck, many word processors (e.g., WordPerfect, Word) now have the ability to create sophisticated web pages.  Ease of construction is the strength.  The greatest weakness is that static content sites are laborious to update as each page must be edited.  To change item descriptions or prices you must edit the pages that contain the items.  If you have a small number of items for sale, and information about those items changes infrequently, the static content approach may be your best web store solution.  

Limited Database Backed Web Store

Imagine being met at the front door of a store by a tireless assistant who's job is to show you the store's products and services that best match product/service requests that you provide.  In response to your query about knitting needles, say, the assistant promptly shows you the knitting needles sold by the store.  Suppose next you want to know what yarns the store carries.  The assistant again springs to life and shows you the various yarns carried in the store.  Even better, the assistant's knowledge is always up to date and flawless.  Impossible?  For humans, maybe, but not for a web site supported by a database that houses product information.

Designing a web store that dynamically presents the products that meet a customer's request involves a linking together web pages and a database. The role of the assistant, in our example above, is assumed by a script, or program, that is responsible for fetching from the database information about the products that meet a customer's expressed interest.  The result is a web site that can provide different content to every visitor, providing each customer with exactly the product/service information that interests them. 

Designing a web site that extracts information stored in a database to meet a shopper's specific request is a task considerably more challenging than the static content approach.  The benefit of a database-backed site is that it is considerably easier to do tasks like add new products, change prices, alter product specifications, etc.  Such changes are made to the database. The new information is then available the next time a customer requests it.

In a database backed site, content and form are separated.  The web store now consists of three primary components:  the look and feel of the site, the site content which is stored in a database, and a script to link the database to customer's product requests.  You design the look of the web pages and you design the database.  Web pages are designed to accept information dynamically extracted from a database.  Suppose, example, you have a knitting store that sells yarn, needles, books, patterns, and accessories.  Each item in inventory would be recorded in a database something like:

Now, suppose that the home page of your web store offers shoppers the option of exploring your offerings in each category.  When the customer clicks on the books category, for example, this action triggers a script that does the following: 

(1)   A message is sent to the server that this customer is interested in books,

(2)   The database is searched and a list of all items assigned to the book category is created. 

(3)   The list of books extracted from the database is inserted into a web page

(4)   The constructed web page is then returned to the customer's web browser. Each listed product is also accompanied by an image of the product.  This is possible because, when designing the database, you had the foresight to include in a link to an image of each product.

A major benefit of separating look and content is that product descriptions, pricing, and other information, is easy to update.  The addition or deletion of products involves nothing more than adding (deleting) products to (from) the database.  The added (deleted) items will then be immediately available to the next customer that requests information about that product. 

The same database can be shared by both retail and web stores yielding only one product database to maintain.  Note that the sample database above includes a 'web' field.  This field can be used to indicate whether an item is to be available only to customers shopping via the web, to retail store customers only, or to both. 

A benefit of the limited database backed web store is that customers only read the inventory database.  Because customers can't change the contents of the database, there are fewer security concerns.

Full Database Backed Web Store

The full database backed web store extends the limited database backed web store by adding real-time inventory management.  Inventory data are added to the product database.  This inventory data is dubbed 'real time' because it is updated automatically each time a customer places an order.  A store's Point of Sale system may be linked to the same database so that quantities available are also updated with each retail transaction.  With real-time inventory a customer knows before ordering whether an item is in stock.  No backorder surprises. 

Real time inventory comes at the cost of complexity.  A full database backed web store is more complicated to develop because each order updates the database to reflect the inventory change.  The retailer always knows exactly how many of each product remains in inventory. 

Cost savings is another benefit of a full database backed web store.  With a web store tied to the inventory database, a number of related processes -- such as picking and packing lists, invoices, accounting systems, and customer data -- can be tied together.  This eliminates possible sources of error and the labor needed to transcribe information from one system to another (order entry and order fulfillment, for example).  The result is a more efficient order processing system. Yet, as economists are fond of saying, there is no free lunch.  All this systems convergence requires expensive systems development and hardware.  Do you need to tie together all of these systems?  Perhaps not.  The following sections discuss alternatives that might be more appropriate for your business.

You can extend the full database web store concept to create customer value by providing additional services.  As we'll elaborate later, customer purchase history data can be used to tell customers about new products that, based on past purchases or past trips to the web store, should be of interest to them.  The customer may also be provided access to his/her purchase history or customer account.  With this access the customer can check on the status of a pending order, track order shipping, remind themselves what they last purchased, etc.  The system could be structured to allow the customer to build a new order from past orders.  But we are jumping ahead of ourselves.  The possibilities for using captured data to enhance to customer value are limited only by your imagination.