Far be it for feeble minded me to critique HBR’s SOA article entitled “Then Next Revolution in Productivity,” ((HBR June 2008, pg 73)) but that’s what I’ll do anyway. While I’m a firm believer that SOA and that we’ll see high levels of SOA adoption in the upcoming years, I’m skeptical that this is the next great leap in productivity. As always, technology is an enabler, but in and of itself does not create productivity. While I’ll say that SOA is a significant evolution in technology, it is not the great revolution compared to the internet or even the personal computer before that for example. So evolution? Yes. Productivity gains? Yes. The next great leap forward? I think not.
SOA will eventually produce enhancements for the business. I hope it’s not like ERP where the promise of great integration and end to end processes cost millions of dollars and may never had fulfilled the promise of its full potential. In the current state, SOA may be so expensive to implement comprehensively (integrating across data, workflow and single end user presentations) that fully functioning SOA is not a reality for any but the largest corporations out there (if even for them). At some point though, price points will come down, point solutions will standardize their integration offerings, and we’ll have everyone down to the mid-market with SOA.
There are some great ideas around SOA however. The top of which is that implementing SOA with only an eye towards integration is the easy stuff, and that data integration is not actually where any significant savings are going to come from. The rewards are going to come from end to end processing (with the assumption that you get good user adoption). This comes from a couple of areas. First of all, the SOA platform offers the possibility of a single integrated user interface that processes workflow and other data from multiple sources in a seamless manner. The user experience becomes a single point of entry and processing that is completely transparent to them. The second is HBR’s idea that organizations must focus on the end to end process itself and apply redesign projects to them prior to implementing SOA. Like any technology project, implementing technology over old processes is simply a duct tape over still broken material. Driving efficient workflow and processes are really where the future rewards will come.
Most of us in HR have multiple applications from multiple vendors these days. We often even have multiple talent solutions from many vendors, which drives our manager end users crazy. Having the ability to provide a single launch point is not enough. We also need to have workflow notifications on that launch page letting managers know if they even need to log into one of the many applications. Better yet, if the data and process that needs to be completed can be brought forward into that page (now a portal I suppose) even better.
SOA will certainly make our lives easier, although I’m not predicting this will be in the immediate future. While it may not revolutionize the way we work, it will certainly allow our end users to see HR processes as less of a headache.