Practical and meaningful enterprise application rationalization is possible and requires good enterprise level modeling that include internal working of the organization, business/data architectures etc. Good modeling is not derived from experience alone but also good theories particularly organization theory and systems theory. I think that the latter is more important than the former since it impacts the direction and the ablility to identify the right things to do. Real life (success or failure) experiences are everywhere but great insights are rare.
Jerry --- Steve Jones <[EMAIL PROTECTED]> wrote: > On 03 Apr 2007 07:07:09 -0700, Anne Thomas Manes > <[EMAIL PROTECTED]> wrote: > > > > > > > > > > > > > > Ahhh... but how much does it cost the company each > year to maintain a bunch of redundant ERP systems? > In many large companies it is measured in hundreds > of millions of dollars. Application consolidation > pays for itself quite quickly. That's why people are > doing it. > > I know, we do quite a lot of it :) But it only pays > in some cases, > from experience the HR, Finance, Accounting and > other "commodity" > pieces tend to work well, while "pure" ERP is often > customised for > good business reasons. I've seen several companies > embark on a "one > ERP" strategy and come really horribly unstuck > because they didn't > recognise what was commodity and could be > standardised on a global > basis and what has optimisations that have real > business value in the > local arena. > > > What I was trying to say is that it really does > depend on the business > and the problem domain > > > > > Organizations spend way to much just keeping the > lights on. They have very little money left to spend > on new projects and innovation. And a significant > portion of every new project is spent on integrating > with the legacy systems. The argument in favor of > application rationalization is academic once you > establish the metrics to measure that cost and value > of your applications. > > 100% agree, but the end result isn't always total > rationalisation. > > > > > IT is broken. The question is how much are you > willing to spend this year to fix it, and how soon > does the company need to see a return on investment > to justify the expenditure. > > > > 100% agree. > > > Anne > > > > > > > > On 03 Apr 2007 02:24:53 -0700, Steve Jones > <[EMAIL PROTECTED]> wrote: > > > > > > > > > > > > > > > > > > > > > Ahhh the vision of "one ERP" there is another > angle here which is "if it ain't broke don't fix > it". The cost of doing an ERP rationalisation > strategy can be measured in the hundreds of > millions, its a nice idea that SAP or Oracle would > do the rationalisation for free but until I see a > case study where that happened I'm not going to > believe that it is possible. Shifting to a global > vanilla (or standard configured) instance is a major > programme of work, both in terms of operations and > in terms of training and "buy-in", treating it as a > pure technology exercise is a sure way to have some > fairly major problems down the road (and ERP is > littered with massive numbers of high cost > failures). > > > > > > So surely the SOA approach is "if it works for > you then fine, we need to wrap it and expose it in a > standardised way so it plays nicely in the > enterprise playground" that way it is up to the > business whether, and when, rationalisation happens > and the rationalisation can be "hidden" from the > rest of the business via business services and their > ultimate technical implementation. Big bang ERP > upgrade is on its way out, the future is incremental > upgrade, and its the business service view that will > enable that to happen. > > > > > > Steve > > > > > > > > > > > > > > > > > > > > > On 02 Apr 2007 09:07:13 -0700, Bill Barr < > [EMAIL PROTECTED]> wrote: > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > The solution is remote integration that > integrates 50 > > > > disparate ERPs in each distributor site. ESB > is best > > > > way to go as of today. > > > > > > > > No, there's a much easier way. Discover which > of those 50 share the same vendor and if there are > 2 or more, call up each vendor and have them come do > the integration/consolidation of their own systems, > free of charge. Then, pit the vendors against each > other for the best and lowest-cost migration path to > one ERP system; the efficiency and effectiveness of > their first-round consolidation will weigh heavily > when selecting the final, sole ERP vendor. Astute > vendor management is a better solution than > technology. > > > > > > > > > > > > -- > > > > email: [EMAIL PROTECTED] > > > > > > > > > > > > > > > > ________________________________ > From: > [email protected] > [mailto:service- > [EMAIL PROTECTED] On Behalf > Of Jerry Zhu > > > > Sent: Monday, April 02, 2007 8:42 AM > > > > To: > [email protected] > > > > Subject: Re: > [service-orientated-architecture] Kaufman on Reuse & > SOA > > > > > > > > > > > > > > > > > > > > > > > > Shared infrastructure... interesting. > > > > > > > > What about a manufacturer having 50 plants > all over > > > > the world distributing their products to > large > > > > retailers such as Home Depot and Lowe's and 60 > > > > independent distributors. There are 50 > ERPs. One > > > > business goal is to have distributors > directly > > > > participating inventory management to achive > real time > > > > order fufilment across the world. > > > > > > > > The solution is remote integration that > integrates 50 > > > > disparate EPRs in each distributor site. ESB > is best > > > > way to go as of today. > > > > > > > > How your shared infrastructure concept fit > into this > > > > scenario? Factor out from 50 ERP systems and > > > > construct one ERP infrastructure hosted in > the Corp > > > > headquater? > > > > > > > > I think that there are three requirements > to have a > > > > shared infrastruture: shared business > process, > > > > Separation of business logic from data logic, > and > > > > separation of content from presentation. > > > > > > > > Jerry > > > > > > > > > > > > --- Todd Biske < [EMAIL PROTECTED]> > wrote: > > > > > > > > > This is a pretty good article, however, they > miss > > > > > out on one > > > > > important factor: Marketing. I still refer > back === message truncated === ____________________________________________________________________________________ Bored stiff? Loosen up... Download and play hundreds of games for free on Yahoo! Games. http://games.yahoo.com/games/front
