Being a longtime enterprise software guy I tend to think of technology architecture when you mention the word “platform” to me. It’s only been in the last couple of months that I started to appreciate what was going on with platforms in our business, starting with Sage and Golden Gate Technology Group, to Oracle, and most recently Salesforce.com. Bear with me as I run through some of these companies and, hopefully, tie it all together to make some sense.
Sage is a pretty well known group out of the UK that has been buying up software companies in the mid-market for quite some time. Dismissed by the larger vendors, they are often referred to as “Microsoft’s problem” in reference to the fact that many of the companies they have acquired have traditionally competed with Microsoft (and Intuit). While acquiring some well known desktop packages like Peachtree and ACCPAC, Sage has also acquired some not so well known vertical solutions like Timberline in the construction vertical. Say what you will about Sage but they are still a billion dollar a year software company and there aren’t too many of those these days.
Golden Gate Capital (and The Gores Group for that matter, but I’m going to focus on Golden Gate for the purpose of this post) is an interesting bird. While primarily a private equity firm, this group has raised a lot of capital and taken Geac, a moderately large public enterprise software company, private and layered on additional acquisitions under the brand name Infor. The bottom line on Infor is that they have acquired a bunch of software products and done very little to innovate, rebuild, or integrate them… their primary focus is leveraging sales cycle efficiencies for capital returns.
Remember ChinaDotCom from the wayback Internet bubble machine? Even they got into the consolidator act in acquiring some well known mid market enterprise technology companies. They haven’t really done anything with these companies aside from consolidating sales and marketing operations and targeting some markets they know very well. In short, it’s a brand strategy.
Oracle’s acquisition spree is well documented and you don’t need me to recap it. However, a curious series of events has happened with them over the last year beginning with their pronouncements that they were going to kill off the acquired apps and convert them to Oracle (I guess this is the equivalent to “kill their leaders and convert them to Christianity”!). Needless to say, that didn’t go over very well and Charles (as in Phillips) and Co. spent the better part of a year reassuring their collective customer base that they weren’t going to kneecap them and that they really intended to rewrite all the acquired apps into Fusion apps and provide an upgrade path, which later became clear that the upgrade path would really be best described as my way or the highway because application extension were going to have to be abandoned. In other words, their customers were going to have to regress their operations to whatever Oracle shipped in the final release, and if you lived through the 11i disaster you were wise to not take a lot of comfort in those statements.
Oracle then shifted gears to “we’re going to wrap all that functionality in Fusion pixie dust and sing kumbaya around the campfire, as if wrapping alone was some magical technology strategy that instantly eliminated all of the problems companies have when they try to integrate applications. As an aside, the whole notion of wrapping code must be one of the biggest frauds that the technology industry as a whole has perpetuated on the marketplace, sure you can wrap it but you pay a performance penalty as you would expect whenever you translate calls and the integration itself is far from predictable because of the proprietary methods that companies use for process error handling and exception conditions, and the UI issues. In short, it never works as well as you hoped and your customers end up pissed off until the point at which you deliver then an integrated application suite.
The most recent Oracle strategy is to abandon wrapping and rewriting in favor of simply maintaining the codebases and separate and distinct products. So much for the acquisition rationale that they started with, that they would achieve significant efficiencies by eliminating the diversity of products in the marketplace for a single Fusion product. What Oracle may finally be admitting, if not explicitly then between the lines, is that their acquisition strategy was really a strategy to control unit economics all along. However, despite my obvious biases against the folks in Redwood Shores, I think this may in fact be their smartest strategy yet because customers increasingly don’t care about the plumbing and the architecture, what they do care about is that the products are mature and that their is a consolidated support infrastructure that they can turn to.
Before I close the chapter on Oracle I want to address Salesforce.com because they have been doing a remarkable thing with their AppExchange initiative. I originally looked at AppExchange as a very clever way to ramp a high throughput partner program by essentially taking the “master” vendor out of the center; in essence abandoning the hub-and-spoke partner model for a mesh network. However, in looking at this lately I am struck by something that I think is far more significant, SFdC is in effect transforming themselves into a platform play without actually delivering any traditional platform technology to their partner marketplace. Of course, all this fit because they are a hosted play and that is really their religion, but when you consider that they are enabling their partner ecosystem with a very simple toolkit (they wouldn’t call it simple, but when you compare it to NetWeaver, Websphere, or Fusion, it certainly is) you can help but admire the elegance of it all.
In each of the above examples, the technology component of the platform equation is not significant, what is meaningful is that significant brands are attaching themselves to a best-of-breed strategy in the realization that their customers are satisfied with a mixed bag approach to applications as long as their is a consistent, stable, and unified support infrastructure underlying them. In some cases the economics of consolidating sales and marketing operations provide additional economic incentive and rationalize marketplace dynamics in favor of the vendors, for the first time in quite a while. Therefore, the emergence of the brand platform strategy is really the realization that our markets are maturing and customers have lessened their reliance on “amazing new technology” in favor or a more risk averse strategy of consolidating the things that work under one vendor. Perhaps the natural extension of this is to realize efficiencies on the build side of the house with component sharing and co-development, but I’ll wait until I see that happening before I actually believe it.