TxM 025 Section 2.8 Comparison between a full-house ERP SEPT re-implementation and a Data Warehouse SEPT implementation Created by James on 7/1/2013 1:58:41 PM
At various points in this document I make reference to re-implementation of your ERP with Strategic Engineered Precision Taxonomies and Configuration and SEPT implementation of your Data Warehouse leaving the ERP largely untouched.
The concept of ERP re-implementation is seductive, most organizations are taking major strategic pain with their ERP's and therefore the logical thing APPEARS to be to re-implement. For quite some time I was seeing this picture and advised a number of clients to go this route.
At the same time, I worked with another client who successfully embarked on a Data Warehouse only implementation.
Then a client of mine, against my recommendations, decided to do a full re-implementation, I bought into the decision and worked with them to develop the project plan based on a fully fledged SEPT Configuration coupled to a fully-fledged systems engineering implementation laboratory as discussed elsewhere in this document.
The client was an engineering firm and the CEO, who wholeheartedly committed to leading the project, was also an engineer. With great thoroughness we set about planning the project and he worked up a full cost analysis. By the time we had allowed for all costs, including temporary staff, cost of senior staff pulled out of the business, comprehensive laboratory testing, etc he arrived at a final budget of around 50 million against my UPPER BOUND estimate of 30 million – I was floored AND seriously embarrassed that I could be so far out.
At the same time, working with another much larger client where the cost of re-implementation would have been in excess of 500 million the penny dropped – the major cost of a full re-implementation of an ERP system involves overhauling and reconfiguring the 80% of the operational and tactical configuration, including workflows (processes) and other mundane components that drive 20% of the value of the investment.
SEPT focuses on the 20% of the investment that delivers 80% of the investment value – BUT, in an ERP re-implementation you have to ALSO do the 80% and so a full SEPT re-implementation of an ERP is likely to be about five times the cost of doing a fully-fledged SEPT Data Warehouse and Business Intelligence implementation against the existing ERP with subsequent trickle down managed by the business!
I have therefore concluded that it is almost impossible to justify a full re-implementation unless the ERP configuration is really so shockingly bad that there really is no alternative but to scrap it and start again. In 22 years of evaluating ERP implementations I have only twice advised a client to scrap the implementation. In all other cases I have advised the client to remediate the implementation and "make it work".
The following tabulation provides comparison of the characteristics of these two alternatives:
|
Item
|
ERP re-implementation
|
New Data Warehouse and BI implementation
|
|
1
|
Trash the existing ERP implementation and configure from scratch as a fully-fledged SEPT ERP implementation project retaining only small components of existing configuration
|
Retain existing ERP, DW and BI implementations and develop a new SEPT Data Warehouse alongside the existing installations
|
|
2
|
Big bang commissioning – I hold the view that it is not viable to commission a fully integrated system module by module and still retain full integration – the installation must be fully established and rigorously tested and trained in the laboratory before going live in the business and most, if not all, modules must be commissioned simultaneously
|
Incremental commissioning of the new SEPT Data Warehouse on an "as required" basis which permits "before" and "after" states to be compared for every element
Trickle the new SEPT taxonomies down into the ERP over a number of years, also on an "as-required" basis
|
|
3
|
High risk project
High risk mitigation cost (laboratory and level of executive involvement)
|
Low risk – in fact, almost no risk
|
|
4
|
All or nothing
|
Incremental "as required"
|
|
5
|
Redo the 80% of the ERP implementation that is working but delivering only 20% of the strategic value in order to be able to redo the 20% that will deliver the 80% of value
Most of the rework is NOT strictly necessary
|
Move directly to the high value investment – the 20% that will deliver 80% of the strategic (thrive) value
|
|
6
|
Strategic information delivery will take years before it commences
|
Strategic information delivery can commence within months
|
|
7
|
HUGE REAL COST
|
Total cost will be of the order of 20% or less of the full cost of re-implementing the ERP
|
|
8
|
Consider only in extreme cases of seriously sub-optimal ERP operation or entirely obsolete legacy systems (then strictly an implementation, not a "re")
|
Virtually every organization of any size will derive huge benefit from this
|
|
>>>
|
There really is no contest
It is almost impossible to justify this option
|
This option will pay for itself many times over in just about any organization of even modest size
|
As you can see, there is a huge case to be made for retaining your existing investment and simply establishing a new instance of your Data Warehouse alongside the existing installations and building an excellent, high value SEPT Data Warehouse and Business Intelligence solution. Or, if you do not currently have a data warehouse to acquire one now...
[MAKERATING]
The comment feature is locked by administrator.
Return