Software as a Service (SaaS) applications have created a lot of excitement and buzz. But combine all this SaaS enthusiasm with the business intelligence goals of many organizations and you’ll quickly find that a major enterprise data integration (EDI) problem looms. The good news is that there are ways to solve this problem — but first, let’s understand the challenge.

The typical Software as a Service environment

Salesforce.com’s SaaS customer relationship management (CRM) software has made it the poster child of SaaS success, with 2007 revenues of $497 million and more than 646,000 subscribers. While there are other software market categories in which SaaS has been successful, CRM and salesforce automation (SFA) have been sweet spots for SaaS for a number of reasons.

First, the end users are typically salespeople, scattered across geographies. These employees generally work in the field, where the customers and prospects are, and not in corporate headquarters. This has made it difficult for the IT group to support sales departments from a software and data perspective. Second, salespeople often keep their information to themselves, on their BlackBerrys or mobile devices, so their sales data is not stored or accessible as an enterprise data asset. Finally, the sales workforce has been ripe for a better CRM/SFA solution for quite a while. The salespeople have needed productivity tools — and their employers have wanted to get that sales data out of BlackBerrys, so they could start using it as an enterprise asset.

So, suppose your company is using SaaS CRM. Great for the salespeople — but there is a major drawback for IT. Although the data is now out of the salesperson’s BlackBerry, it is still not within your firewall or in your enterprise data warehouse. Your SaaS application has just become another data silo!

Enterprise data integration vendors rush to offer Software as a Service tools

A few enterprise data integration vendors have developed new technologies to mitigate the problem. Vendors such as Redwood City, Calif.-based Informatica Corp., Oracle Corp. (via its Sunopsis acquisition) and Austin, Texas-based Pervasive Software Inc. have developed new tools and services to help companies extract data locked in the SaaS data silos.

The data integration capabilities vary by vendor, but often include the initial data migration onto the SaaS application; pulling data out of the SaaS application and into the enterprise environment (such as into a data warehouse); and finally, data integration or exchange between an enterprise’s off-premise and on-premise applications.

When you evaluate these vendors’ products and services, you need to understand what’s available and what style will meet your needs. In order to enable data integration to and from off-premise applications, a data integration product could:

• Offer a connector to the off-premise application that could be used with your on-premise data integration tool. You design and deploy your data integration scheme on-premise and access the off-premise application as just another source/target (that happens to be outside your firewall).
• Offer the data-integration capability as an on-demand (off-premise) service. In this case your data integration functions and your SaaS application are both off-premise and you would use the on-demand data-integration service to bring data on-premise into your enterprise, so you could integrate it with other enterprise data.

This all may sound elementary, but many enterprises start off with just thinking about the SaaS application and its capabilities, without thinking about data integration or business intelligence and reporting. This common oversight shouldn’t come as a surprise to many people. It is often exactly what happens when on-premise enterprise resource planning applications are initially deployed — reporting and data integration are afterthoughts. I am constantly reminding my clients that, when deploying new enterprise applications, they need to incorporate that application’s data in their overall enterprise information architecture.

So, whether the SaaS application is CRM or something else — just make sure that SaaS doesn’t contribute yet another data silo to your organization.