There have been several articles written about CEVAs but they all seem to have been written for the customer not the vendor. I think it’s too early to talk about how a customer should choose a CEVA when there’s only one CEVA to choose from. What we really need is for vendors and VARS to look at a CEVA strategy as a new way to build their products.
It’s simple to say that we should relate CEVAs to relational database platforms, but even this level of clarity doesn’t exist for CEVA. There are basic questions around CEVAs that have not been answered. So let’s answer them.
ECM Platforms are Overkill
One of the first statements critics make is that ECM platforms are overkill. My first response is, so what. I wonder if these people also complain when they are offered a free desert with dinner. Yes, today most CEVAs would only use 25% of the features an ECM platform has to offer. But the level of effort to develop that 25% is not that easy.
The minimum feature set an applications requiring content would us is library services, file locking and versioning to ensure integrity and security to control access. Building these services from scratch requires at least two or three developers six to nine months to design, develop, and test. Once in production the functionality is unique enough that it would require some dedicated time for customer support. While it may seem easy, code lines rarely go bug free nor does a vendor stop at version 1.0. This means you would have four people dedicated to developing and maintaining library services year round.
And let’s say you developed library services in house, what’s next? Customers will want to add transformation services to support PDF renditions. Well add another developer. Now your customers need simple retention services. Add yet another developer. If you think I’m wrong just look at the team that John Newton at Alfresco had to pull together to develop a competing platform in three years. Over time your resource commitment will continue to grow. To parrot our parents, “You’ll grow into your ECM Platform.”
ECM Platforms are to Expensive
Well here I would tend to agree with you. If you look at the base level of features library services have offered for the past ten years nothings changed. One hundred dollars still gets you the same basic library services, everything else costs extra. When you talk to an ECM vendor as a VAR (value added reseller) they see you as a direct source of revenue rather than a pathway towards additional customer revenue, and want to charge through you the same dollar value for the basic library services, even if you bring ten times the functionality to the table.
If you talk to any of the major platform vendors, they want you to put your product’s technical future at risk by building on their platform and they want you to pony up a wad of cash for even the privilege to try. For this lump payment they offer no additional services to make platform enablement easier. It’s not surprising that when their platform strategy fails they blame the customer rather than themselves for their lack of success. They really don’t recognize the fact that working with a VAR means working together.
Instead of choosing a major vendor, work with open source solution, like Alfresco or Joomla, as the platform for your CEVA. You get a solid platform to build on and a vendor or community looking for ways to increase customer adoption. Using open source also means that the additional software costs are kept to a minimum ensuring that you get a majority of the reward for value that you are adding. The one drawback is that you will need to support it yourself, but you would have had to support your own ECM platform if you had built from scratch.
It won’t take long that the attention your product starts to get from customers becomes visible to the major ECM vendors. Ultimately ECM vendors will work with CEVA vendors, just look at the pricing strategy from Oracle.
Which ECM Platform to Choose
It’s only in science fictions movies that “there can be only one.” Take a hint from all of the ECM vendors. Most ECM systems work on both Windows and UNIX platforms and with both Oracle and SQL Server. Why should your CEVA strategy be any different? CEVAs should be vendor agnostic and don’t let your ECM vendor tell you any different. And if you’re about to say it will cost me more, remember you were thinking of building your own library services. Building an integration with two or three ECM will still take less time and money then building your own. And it gives you a larger market base for your solution. But which platforms to choose?
You need to start by looking at the problem in the industry you are looking to solve. Find out which ECM platform is being used by your customers and try to work with them at least and one other. For instance in the legal space you should probably look to work with Open Text or in pharmaceuticals you would work with Documentum. Then find another vendor that’s hungry, open sources fit this bill. So you go in with the lead vendor but due to costs you also work with the open source solution. Not only do you get to platforms to hedge you but on you also get a “Plan B” for sales opportunity.
Building the integrations to different ECMs can be easy. Key is to start by normalizing the calls into a series of functions that you call in your own application. Never call their APIs directly from within your logic. Once you start calling the ECM’s APIs directly you’re locked into their platform.
One Major Point to Consider
Most of the questions asked when considering CEVAs can be answered quickly. Are ECM platforms overkill? This may be the case today but in the long run their features offer growth potential. Are ECM platforms to expensive for me to use? Some vendors are but there are other vendors and platforms that can achieve the same results. Which ECM platform to choose? Be Switzerland and work with at least two. But their one major point that needs to be considered.
In building a solution to solve a business problem, you have limited resources. Just as you wouldn’t build your own database platform from scratch, why would you even consider building an ECM? By building on an existing ECM platform you can use your limited resources to build additional business value and limit some of your risk. In the end your market opportunity today is probably limited by another vendor but against your prospects building their own.
Click for more about OEM’ing an ECM Platform.
Your point about ECM platforms being overkill for individual applications that may only use 25% of the platform’s functionality is fair. I would add that although one CEVA may use one 25%, another CEVA will likely use a different or partially overlapping 25%. Take Accounts Payable and Legal Document Management as two potential CEVAs applicable to most enterprises. The AP application will likely focus on image handling, large volume ingestion, and load-balancing workflow, while the Legal application will likely focus on collaborative authoring, a more complex taxonomy and fine-grained access control. The latest generation of ECM platforms can handle both these use cases, but both need quite different applications and user-interfaces.
I agree that the benefits to the ISV may not be immediately clear, and certainly there is a loss of control in developing on another vendor’s platform. But I really don’t see much choice. It is the customers that are driving the market in this direction, because the benefits to them of standardizing on an ECM platform are clear: reduced support costs, improved information sharing, and above all, the potential to mitigate the risks created by the “Diverse and Exploding Digital Universe” (to borrow the phrase from the latest EMC/IDC study). Creating more siloes of content is not the answer.
You’re absolutely right my 25% may not be your 25%; it is all about the application. A more extreme example would be managing engineering CAD drawings.
The ECM vendors themselves have used the CEVA model for some time GMPharma from Documentum and LegalKEY from Open Text, but the real advantage of the CEVA model is when those outside use it. The vendors unfortunately don’t see it this way and instead try to squeeze the value out of those looking to use their platform.
I’m slowly trying to build a list of vendors that are successful in developing CEVA applications. I’m ignoring those by the vendor or those that are administrative utilities for an ECM platform.
It interesting to see that applications like Altien for Corporate Legal (built on FileNet), Prodagio Contracts and A/P from Imagitek (built on Documentum), and HotBanana (built on ColdFusion CMS) have decided to take the leap into CEVAs. Hopefully their success can drive others to do the same. The Prodagio applications were recognized as Offering of the Year from EMC for both 2006 and 2007, an award which is base on driving revenue or more appropriately showing traction.