Saturday, November 04, 2006

A New Era for ECM

I think that the second half of 2006 will come to be viewed as a tipping point for the ECM market. Analysts have been predicting vendor consolidation for many years, but it just didn’t happen. That is now changing fast. IBM’s acquisition of FileNet, Open Text’s acquisition of Hummingbird, and last week’s news that Oracle will acquire Stellent signal that the market is shifting into a new phase.

John Mancini, President of AIIM, rightly points out that the ECM industry is following a familiar economic pattern of consolidation, and that we are now entering the final stage when a handful of players achieve lasting dominance. So what will the market look like in the future? Economics tells us that three or four dominant players will likely take at least 80% of market. Many commentators have drawn parallels between the ECM market and the relational database market, and I agree that this is a valid comparison.

According to Gartner, in 2005 80% of the RDBMS market was shared by three vendors – Oracle, IBM and Microsoft. In 2005 80% of the ECM market was shared between eleven vendors. Following the closure of the deals mentioned above, the number is down to nine (Oracle wasn’t in the top eleven). Using last year’s numbers, IBM + FileNet is now the clear market leader with ~26% share, OpenText + Hummingbird is second with ~20%, EMC is third with ~13% and the rest (Interwoven, Vignette, Oracle/Stellent, Mobius, Hyland and Microsoft) are in mid to low single digits.

Of course the relationship between the ECM market and RDBMS market is more complicated than a simple analogy. At one level, today’s ECM server engines are just relational database applications, albeit some of the most sophisticated such applications in the enterprise landscape. Alan Pelz-Sharpe wrote an excellent piece earlier this year on CMSWatch about the future of ECM in which he supported the opinion that “the database vendors will in future own the repository”. The database vendors undoubtedly have a technical and marketing advantage. IBM and Oracle have already shown through their acquisitions that they getting serious about trying to dominate the market. Microsoft is finally getting serious about ECM with the launch of Office 2007 and Microsoft Office SharePoint Server 2007. Also, the decision to pull the long (long) anticipated relational file system WinFS out of the desktop operating system, and to repurpose the core technology into a future release of SQL Server, shows where their long term vision lies.

So it seems probable that we will end up with three or four dominant ECM platforms. Does this mean the end of innovation in ECM? I think the answer is absolutely the opposite.

John Newton of Alfresco wrote very eloquently in September about the Commoditization of ECM. He argues that despite many false starts, standardization will now come to ECM, largely because it is now in the interests of the major platform players to accelerate it.

“The scene is set for real standardization in content management and commoditization to the point of real replacement and swap out of existing systems. The environment is very similar to that found when the database market first standardized through SQL-89 and SQL-92. This was basis upon which the smaller players in DBMS disappeared and the new era of client-server and ultimately the web appeared. The changes in the ECM market can be just as profound.”

As an aside I would say that Alfresco has an excellent chance to emerge as a major ECM platform player. The open source model had not evolved at the time that the RDBMS market entered its end-game phase. Despite this MySQL has emerged as a significant force. Alfresco is well positioned as the ECM market is just entering the end-game and it will be fascinating to see how it disrupts the market.

Standards such as JSR-170, its upcoming successor JSR-283, and the iECM initiative from AIIM, promise to deliver true interoperability across ECM platforms. It is my belief that this process of standardization, together with the ongoing process of platform-vendor consolidation will actually enable a flourishing of new content-centric applications, which will transform the market landscape.

Why? Well to start with, the market opportunity remains enormous. Analysts frequently state that 80% of the world’s data is unstructured and that only 5-10% of that unstructured content resides today in a content management system. Last year the RDBMS market was worth ~$13bn while the ECM market was ~$2.5bn. I know several customers that have a vision for “total ECM” but none that have actually delivered it. That gap will only be closed by new applications that address the myriad use-cases around unstructured content. It is inconceivable that a single vendor could address this entire market effectively: a single platform yes, all the applications no.

The second point is that until now it has been nigh impossible for a developer of content-centric applications to address the market effectively. The lack of standards and the fragmented market meant that supporting one or maybe two ECM platforms was the limit. The addressable market for a new content-centric application was thus pretty limited, and the ISV had the added headache of hoping they had backed the right platform horse. In the new world, an ISV can write their application once against an interface standard, or if they need platform functionality beyond the scope of the standard they will only need to work with a handful of platforms. At Altien we have already solved this problem by integrating ADM with IBM WebSphere® Information Integrator Content Edition (IICE) which provides a fully-functional bi-directional interface to all major ECM platforms.

And this issue looks pretty similar from a customer perspective. There must be a lot of anxiety right now among customers about what consolidation will mean - especially where they have built customized applications onto what might now become a defunct repository. Large software vendors are generally good at talking about supporting acquired platforms – but in practice, where there is overlap, one platform will be quietly starved of R&D, despite what the marketing rhetoric might say. Given all the uncertainty now about which platforms will survive, customers must be asking how they can best preserve their development investments. The only real answer is for customers to start building ECM applications that are platform independent. With IICE, IBM has a unique piece of middleware that enables their customers to adopt this strategy today. For Hummingbird and Stellent users, the picture is less clear.

So what will these new content-centric applications look like? I think the scope is enormous but verticalized, business process centric applications looks like a big opportunity. I saw Forrester analyst Connie Moore present at the IBM IOD conference and one of her points was that the current major ECM players that weren’t going to win the platform battle would need to verticalize to survive.

I see another key trend converging with the ECM market in the form of Office 2.0. I recently attended the hugely positive Office 2.0 conference organized by Ismael Ghalimi of Intalio. In his wrap up blog post, Ismael refined his definition of Office 2.0 as follows:

"Office 2.0: Office productivity environment enabled by online services used through a Web browser. By storing data online and relying on applications provided as Web services, it fosters collaboration and extends mobility, while promoting a user-centric model that fuels innovation and increases productivity."

What I saw was a host of new companies that are trying to change the model of personal productivity applications that we have lived with for the past 20 years. Some of these applications are about new means of creating content, some are about new means of organizing, sharing and collaborating on content, and the best are using the technologies of the web to combine all of the above.

So are these applications ready for the enterprise? Ismael asks the question "Who is it for?" and answers as follows:

"Individual users and Very Small Businesses (VSB). On one hand, Office 2.0 is not ready for the enterprise, and it’s a good thing. Trying to make Office 2.0 work for the enterprise today will strip the concept off all the good things that make it interesting. The enterprise needs reliability, scalability and security, and such attributes take some time to implement."

He goes on to the make the valid point that Individual users and Very Small Businesses (VSB) represent a huge market opportunity, but ironically much of the debate at the conference was about how to penetrate the enterprise market.

To the attributes reliability, scalability and security I would add integration as a core requirement for enterprise-readiness. Large enterprises today are wrestling with the problems posed by the exponential growth of unstructured content and its fragmentation until multiple silos. The last thing they want is more silos. The integration of Office 2.0 applications into an ECM infrastructure would enable the delivery of all the enterprise-readiness criteria listed above.

And I think the enterprises are ready for these solutions. For a start, large enterprises have moved overwhelmingly in favour of web applications for almost all general business applications. And if you have a few thousand desktops to manage you know why. And second, knowledge worker productivity seems to be being thwarted by the current model, competition for talent is increasing, and organizations want to get more from their most valuable employees.

In summary, I am very optimistic about the prospects for ECM. Rather than representing a simple loss of choice for customers, ECM platform consolidation will provide new choices as it enables the development of a new generation of content-centric enterprise applications. We can already see glimpses of how these applications could dramatically transform knowledge worker productivity. I expect to see a wave of innovation in this market in the next few years. This is an important trend and one in which Altien will play its part.