I recently spoke to Ram Gupta, the new president of Cast Iron. Gupta is very bullish about the future of application appliances. In fact, he came out of retirement to take the helm of Cast Iron. Gupta was previously an EVP at PeopleSoft, and said while there he spoke to hundreds of CIOs who all said integration was their number one pain point.
Gupta says that out of $100 billion being spent on integration each year (his calculated number) 80% go to people, and most integrations are point-to-point. He states that the problem is most integration middleware solutions are too complex. Cast Iron is trying to commoditize the integration market with the integration router, similar to the way Cisco commoditized the networking market with the router (does anyone remember writing to network protocol stacks?).
The Cast Iron integration router is positioned as a fast an easy way to connect two applications. Gupta believes simple point-to-point integrations represent 70-80% of the market. Cast Iron plans to make it even easier to connect applications by providing application specific routers, for example, a SalesForce specific router, with all the data mappings already done. The goal is to make it possible to integrate an application in one day. The concept of "packaged integrations" is not new. CrossWorlds attempted to do the same thing in the early days of integration. SeeBeyond followed and offered its own application-to-application specific solutions. None of these early attempts were particularly successful.
I concede there is a real and urgent need for simple solutions to simple problems. Integration routers will have broad appeal to different size companies. Large organizations will adopt them as part of their integration arsenal for fast tactical solutions. Mid-sized companies and branch offices may find it solves many but perhaps not all of the integration requirements. It is extremely suitable for small companies and satellite offices that have no IT staffs. I do in fact believe there is a significant potential market for integration appliances. However, Cast Iron is not alone in this market. Cisco is introducing Application Oriented Networking (AON). Gupta insists that Cisco is not a competitor, but I remain unconvinced.
Despite the fact that it is clear Ram Gupta believes he's about to ride another rocket of company growth, I'm afraid I need to burst his bubble and significantly deflate his estimate of potential market share. I do not believe that 70-80% of integration technology investment will be spent on routers. Composite applications and SOA, business process optimization, and business activity monitoring with predictive capabilities are the leading edge capabilities many companies are seeking. These capabilities are beyond the purview of application routers, are far more strategic to the organization, and have a much higher ROI potential. I believe these are the areas in which companies need to be investing most heavily.
Where is your company investing IT dollars?
Beth, you raise a valid point that more complex products are necessary for more robust needs but we've been very happy with our appliances as a way of completing integration projects (file, ERP, database, etc) faster with fewer people.
Brian, thanks for your comment. I totally agree that integration appliances have their place in the organization (in all size organizations). My objective was to show where they might fit, and where they don't.