Implementing an ERP is a tall task, you may engage an integrate or tear down operations
As I was stating on the previous article, implementing an ERP requires a clear vision of where we are heading; what our current processes are and how we want to move forward, this lead us to answer several questions like:
How we do it?
How we want to do it?
where do we have redundancies?
what are the links between activities?
who does what?
how is the delegation of authority designed?
One thing is to implement an ERP on a single company, and a whole other is to do it within a frame of several companies, on top of that, difficulty increases when it comprehends several countries. We all know there will always be resistance to change, but sharing best practices is the way to improve processes and the baseline to implement an ERP. I remember while I was working at a swiss company, we started the project of implementing SAP, I was a global key user and local BPO for a couple of modules and also, as finance responsible, I was involved on the design of several other modules. The ambitious project cost around 40 million swiss francs; the intend was to roll out SAP in 20+ countries with Mexico, US and Canada as the tipping point. Nonetheless we needed to double check with countries as different as Germany, India, Brazil, Southafrica, among others, looking for the best practices. You can't be naive and think you can implement the same ERP worldwide as long as each country has different regulations, tax laws, and so on and so forth, so we needed to have tropicalizations in place, but at the same time trying to minimize them up to a 15 to 20% of the total design.
So it begun, drawing all the flow charts for every single process in place, and then start drawing the "desired processes", it took close to 3 years to compile, design, discuss , compromise and finally agree on the desired state, I still remember the so called blueprint, including each flow chart each table, each DoA criteria; it took more than 1,000 pages put together to get this blueprint done. One thing I must not forget to mention, is that it also had the whole "corporate" structure, which was a hyerarchy built over legal entities, profit and cost centers, production planning, workorders, etcetera. We also created parallel ledgers which had not only local GAAP, but reflected on different sets of books IFRS and tax accounting for each country; we threw in cost elements, business units, the whole nine yards; everything under the umbrella of a single, lean , neat chart of accounts shared across these 20+ countries. The total project lasted around 4 years, uncountable hours, meetings, conference calls, travel time, testing periods, trainning sessions.
The weeks prior to go-live were stressfull. Getting all the legacy data in place, reconciliating whatever was going to be uploaded against legacy systems, the goal was to avoid keeping parallel systems as we went live, therefore whe could not by any means have design or conceptual issues ir else we were going to stop the whole operation and information systems of 3 countries; a Sunday evening we uploaded all the information, in that precise moment Navision (our previous ERP) died and SAP switch was turned on. In all fairness we had a few bumps, but nothing major, they were more training related than design or implementation problems, so that day in august we succesfully launched SAP into our organization.
But sometimes implementations are neither as thorough, nor as detailed as we did, a few years back I found SAP on another american company and Oracle on a major mexican company, what I found is that in the latter the implementation kept the old structures, subledgers were not integrated correctly and the hierarchies were not defined, therefore instead of cost elements to segregate data, we were stuck with 8 segments within the chart of accounts (as old concept as my first implementation on 1990), so there was no matrix on data and everything needed to be coded to those specific segments which complicated information systems, also there was no data warehouse neither Business Intelligence, so in order to have a top to bottom report you needed to built it bottom to top, and it was a messy way of getting data, because instead of being able to drill into the information you needed to brick it up to get it; the biggest hurdle was the lack of licensing, which is ridiculous for a company of that size and the MIS was so lame you couldn't get info in real time nor accurate information until it was a little bit too late to prevent deviations. The worst part of this implementation was that even though we knew it was not properly done, instead of "making a major surgery" to correct it once for all, we used to take aspirins and try to "fix it" one step at a time; I understand that sometimes throwing away a multimillion investment may not be the wisest move, but if you measure the time and effort wasted by the whole organization just to put together information and then arrive to different conclusions because of poor acounting methods and unefficient information systems, I wonder which is higher, on top of that due to the poor information systems we required additional workforce all accross the board just to be able to put info together.
I'm conviced that implementing information systems whether it's an ERP, CRM, or any other kind, the coneptualization comes first, then hyou have to assess which platform will provide whatever you are dreaming of, and afterwards the design of the desired status, definitely any implementation requires more planning time than programming it, if you are doing it backwards you may probably end up with the hot potato.
Comentários