Managing the evolution of a mature software product is not easy. My previous posting lamented how Microsoft had made a mess out of Office, perhaps the most profitable franchise in software history, by overloading it with esoteric features. Thinking about the Office tragedy has led me to ruminate on how the same dynamics could damage our favorite ERP suite by making it unnecessarily complex.
JD Edwards customers tend to be more passionate and loyal about it than the users of most other ERP suites. One of many reasons is that both World and EnterpriseOne have matured into comfortably usable products. There isn’t an awful lot that typical users want them to do that is not already there. As a result, each time the design team looks for suggestions as to what to put into the next release the requests that arise become increasingly narrowly focused or esoteric.
In the Collaborate sessions where ideas for improvement are solicited I am always surprised by how few suggestions of real substance are made. Almost none are fixes of serious flaws or obvious shortcomings. I am not suggesting that the JDE product line is anywhere near perfect, only that it has reached a stage of maturity where the obvious rough edges have largely been sanded away. The few serious limitations that do exist are generally covered by partner offerings such as the DSI bar coding solution or my own company’s pre-built data warehouse.
From a user’s point of view, it is a good thing for a software package to be mature and stable. The problem comes when the annual maintenance bill arrives and management wants to know what it is paying for. In many cases, hundreds of thousands of dollars per year are involved. Cranky individuals in the finance department increasingly want to know exactly where the value comes. Part of the justification comes from the very valuable support that Oracle provides including bug fixes, help when problems arise, and routine updates such as tax tables. This alone, however, does not justify the full cost of software maintenance.
When customers first buy software most are convinced to signup for annual maintenance in order to get access to the improvements that come with new releases. In the early stages of a product life cycle the improvements tend to be dramatic. They are also invaluable when responding to technology change such as the emergence of the web.
The current dilemma for the Oracle executives that serve our needs is that the basic JDE products are mature enough that there is not a lot left to put into new releases. As a result, much of what is in the improvement pipeline are nice to have features that are only relevant to a small fraction of the community (e.g. fashion industry enhancements to handle sizes, colors and collections).
Lyle Ekdahl and the other Oracle executives that plan JDE releases should be commended for not falling into the trap that Microsoft did with Office 2007 – overloading a good product with so many esoteric features it became unusable. Taking this minimalist approach, however, is risky since it can bring into question the value of continuing to pay for maintenance.
The issue is not that Oracle is failing to invest heavily in valuable new capabilities. The problem is that most of that investment is in products that fall outside the JDE applications suite. Examples include Fusion Applications, edge applications such as Demand Planning, Business Intelligence, and in hardware/software appliances. These investments have the potential to create a great deal of value for us in the long run.
It requires an act of faith to pay the current high price for maintenance in order to get undefined benefits in the future. The advice my company gives to most of our JDE clients is to grumble if it makes you feel better and then pay up. Dropping maintenance or going to a third party pays off only if you hold off on upgrades for a long time. Our best guess is that Oracle will provide enticing new capabilities outside the JDE product suite in a way that will reward those on newer releases.
Oracle (and its competitors) charge a great deal for software maintenance because they can. Expect Oracle to continue to do just enough to keep most of us paying. While this approach is not ideal, it is better than the Microsoft Office strategy of overloading a good product with too many new features.
If you face the dilemma of whether to stay on Oracle maintenance for JDE applications we would welcome the opportunity to share more thoughts with you individually. Anyone willing to share opinions on this important topic with the whole community is encouraged to submit a comment.
May 19, 2010 at 4:16 pm
Very relevant and nicely written post. One area where Oracle will be providing significant value is in FMW enterprise content management inyegrations with Oracle Forms Recognition and OIPM inyegrations for automating paper invoice processing.
May 20, 2010 at 8:12 am
Thanks David for your informative posts and I do agree with you that JD Edwards E1 is a mature full featured product. Yet coming to this suite with fresh eyes, I can say that there is at least one area that needs serious attention; that of the developer’s experience. There is lots that should be done to reduce implementation, integration and development costs. We’re still dealing with antiquated developer tools; when will we see revision control for the source code? Then there’s the poorly documented business functions (you’re lucky if there’s any documentation at all). When you’re dealing with over 10,000 business functions and over 20,000 methods, clear documentation is a must. While I haven’t fully delved into the available web services what I read in the documentation and heard from others leads me to believe that the list of available services is short and those that exist have many limitations.
There’s lots more, I wonder what else others might like to see?
May 20, 2010 at 9:49 am
We went off of maintenance 6 years ago due to escalating yearly maintenance costs. We were on World 7.3 and was very stable so we did not have support calls so was basically paying for 1099 and W2 updates. The future of JDE was unclear to me at the time because Fusion was just a theory and there was no commitment from Oracle to enhance the World product at the time. Any enhancements we needed we just wrote ourselves or integrated a 3rd party package. We have upgraded from OS V4.5 to V5.2 to V5.4 but now we would like to go to V6.1 but we cannot because all objects must be observable. We paid for JDE source but do not have all of the source and Oracle will not give those to me. Anyone else in the same situation that I am in? What are your plans?
May 20, 2010 at 1:31 pm
Shadrach,
You make a good point – integrations with other applications will remain a very important source of enhancements for many years to come. Such enhancements do not bloat the product or add significantly to its complexity but can add a great deal of value. They represent another reason to consider staying on maintenance.
Dave Andrews
Editor
May 20, 2010 at 1:39 pm
Ron,
Thanks for the thoughtful comment. I was primarily thinking of the applications themselves and not the development environment when I wrote this posting. My own coding days were a very long time ago so I am less qualified to comment on what development improvements are needed. My sense is that there is more work left to do there than on the application side partly because the art of software development continues to improve faster than that of business processes.
Your posting already has caught the eye of some of the JDE Denver people. If they come up with a worthwhile response to the concerns you raise I will make sure it gets back to you.
Dave Andrews
Editor
June 7, 2010 at 12:24 pm
[…] How maintenance revenue can lead to feature bloat « The JD Edwards … […]
August 3, 2010 at 7:26 pm
Comment for Bill and those on 7.3 version
We are also in similar situation with a heavily customised 7.3 version. Support due to expire in 2013 and we need to start researching our options. So far, any integration options hit a wall due to our customised version and the fact that most of the code is sitting in applications and not in BFs. Integration seems too dificult (maybe due to limited JDE developer expertise in Australia??). Anyone who has upgraded and/or integrated (say with BizTalk or similar) the same version of JDE I would love to hear about your experience.