… provided, as this article on SaaS Integration in Intelligent Enterprises astutely notes, it’s a forethought and not an afterthought. As the author notes, SaaS, like any other software delivery model, is still a sub-pattern of enterprise architecture. This means that you have to account for the architectural requirements in your selection of a SaaS solution. Just like it’s hard to fit a square peg in a round hole, it’s hard to fit a SaaS solution that only communicates in XML with an old on-premise application that only communicates in EDI (unless you have a middleware mapping tool and are prepared to hire an expert architect who understands both products and can define the appropriate mappings for you).
Since the ultimate goal of your supply chain organization is to put together a “suite” of applications that allow you to analyze, monitor, and execute the supply chain end-to-end in a seamless fashion, you need to insure all of your applications synch up through a central data store (which could be on the Cloud). Thus, you have to insure that any SaaS product you buy is capable of exporting and importing the required data from your central data store in one of the formats you can support. You also need to insure that the mappings and integration to other modules that the application needs to synch with can be done in a reasonable time frame at a reasonable price. And if you do this work up front, integration will not be a challenge. In fact, since most modern SaaS applications were built with the understanding that they will have to suck data in from internal sources and spit processed data back out again, as long as you select the right SaaS application for your needs, integration won’t be hard at all.