Marketing software has evolved over the past decade from stand-alone systems for database development, segmentation, campaign management, reporting and analysis toward integrated marketing suites. But while integration seems inevitable, the outcome of this process is uncertain. What remains unknown is exactly how the integration will occur.
Siebel Marketing Solutions (Siebel Systems, 650/295-5000, www.siebel.com) represents one vendor’s approach to integration. Siebel dominates the market for high-end customer service and sales automation systems. Its Marketing Solutions comprise three modules – for campaign management, data analysis and Internet marketing – that incrementally extend its existing products. This is a sharp contrast to vendors with their roots in outbound campaign management and analysis, whose idea of an integrated marketing solution neatly excludes the operational contact systems themselves.
In practical terms, Siebel’s approach means that its marketing solution is centered on the same customer database as its operational systems. This statement should alarm every veteran of the database marketing industry, which spent much of the past decade arguing that marketing and operational databases should be physically separate. The difference is that marketing databases read data on many customers at once, while operational databases write data on one customer at a time.
Efficiently executing both functions in the same physical structure is generally considered impossible (though a few systems, notably NCR’s Teradata database, make a credible claim to the contrary). The good news is that Siebel has not made this mistake. The analytical component of its marketing solution is built on star schema data structures, which are the approved structures for classical marketing databases.
Data are periodically extracted from Siebel’s operational databases and loaded into the stars, which are used for analysis and selections. This portion of the system is so distinct from the rest of Siebel that the vendor has chosen to build it with third-party software: Informatica for data extracts and transformations; and Business Objects for reporting and analysis.
So far, so good, though the use of third-party software weakens the integration benefits from Siebel’s solution. (Actually, Siebel claims only that it has simplified implementation by pre-integrating the different pieces of its solution, which is true. The final changes necessary to complete a real-world installation still must be made in the three separate systems.)
But Siebel has not just built an analytical database extracted from its operational systems. The heart of Siebel Marketing is its ability to link analytical and operational systems so the distinction largely disappears from the user’s view. At the data level, Siebel does this through integration technology acquired in its January 2000 purchase of Paragren Technologies.
This technology lets users define customer segments using a combination of operational and analytical data, as well as external data that have been mapped into the system via open database connectivity connections. In practice, selections are made first against the analytical data – these tend to involve the complicated queries that analytical databases are good at.
The resulting list of customer IDs is then passed to the operational database, where names and addresses are appended and simpler selections are made using the up-to-the-minute information. Because the last step in the selection process occurs in the operational database, Siebel can record the selections in its main contact history file, thereby ensuring that a single, comprehensive contact record is immediately available to all operational systems.
The complexity of this process is largely hidden from the end-user, who instead sees a reasonably powerful, reasonably easy campaign management interface. Marketing programs are laid out on a flow chart that lets users include multiple customer segments and outside lists, allocate names among multiple offers and execute multiple steps over time. This should be adequate for most purposes, though sophisticated database marketers may have complaints:
Complex selections rely on an administrator or advanced user to pre-define calculated measures; large numbers of segments are awkward to manage in a flow chart; and the interval between program stages must be built into segment definitions rather than defined explicitly on the flow chart itself. But other portions of the system are quite advanced. It captures cost and revenue estimates for pro forma calculations; links each contact to offers listed in a separate database; lets users create complex source code definitions and export formats; and can apply standard filters with separate inclusion and exclusion logic across selections.
But while competent campaign management is important, the real advantage of the system is its close relationship with other Siebel components. At the functional level this means it has access to standard Siebel functions, including a scheduler to run campaigns on a regular basis, fine-grained security to control user access to different data elements and capabilities, and work flow to manage the tasks involved in building and approving a marketing project. More importantly, campaigns designed in Siebel Marketing can be directly deployed to operational systems to execute in real time – which requires complicated custom integration in most other marketing systems. This could be an overwhelming consideration for marketers struggling to coordinate outbound and inbound customer contacts. Of course, it only works if they use Siebel operational systems as well.
In addition to campaign management (Siebel Marketing) and analytics (Siebel eBusiness Analytics), Siebel Marketing Solutions also includes Siebel eMarketing, which generates personalized e-mail messages, newsletters, surveys and Web pages. Like the campaign manager, this is a respectable tool whose primary advantage lies in its close ties to other Siebel functions. For example, Web surveys are written in the same scripting language as telephone surveys, so they can easily be deployed in both places and can post replies directly to the main customer database.
The current release of Siebel Marketing Solutions runs on Windows workstations and Windows NT servers, although a browser-based version is due in the first quarter this year. Marketing and eMarketing are priced on the number of users and database size and start at $250,000. The cost of eBusiness Analytics depends on the reporting modules and number of users and starts at about $200,000. Components of Siebel Marketing Solutions have been sold to about 60 firms.
• David M. Raab is a partner at Raab Associates, Chappaqua, NY, a consultancy specializing in marketing technology evaluation. His e-mail address is [email protected]