Features - Enterprise Data Insights:

ERP II - DÉJÀ VU?
By Andy Hayler, Founder and Chief Strategist, Kalido

Large companies the world over rushed to get their ERP implementations completed to beat the Y2K deadline. A bizarre trend is now emerging: the re-implementation of ERP, otherwise known as ERP II. Many large corporations found that, having spent huge sums of money on ERP on the promise of "standard business processes for the measurement of overall business performance" they didn't quite manage it the first time round. Instead of having different versions of their corporate data locked away in different systems, now they have different versions of their data locked away in multiple, incompatible ERP implementations. Will the money for ERP II be any better spent this time around?

It would be interesting to re-examine those former cost benefit cases for ERP: "Ah yes, we can justify this vast expenditure on the standardization of our business processes, which will have great savings in the business. And of course our IT budget will be greatly reduced as many of our old systems will be replaced by a new package". Well, if the holy grail of standardized business processes did not occur, where are the benefits? Also, has anyone noticed their overall IT support costs dropping massively after an ERP implementation?

One might imagine a certain amount of chagrin on the part of those managers who recommended the ERP implementation, given that the benefits have proved so elusive. Apparently not. Figuring that the best means of defense is attack, project proposals are now being mooted to "re-implement" the ERP systems in the form of ERP II. Why it is felt any more likely that business processes can be bludgeoned into standardization this time around, when they manifestly did not last time, is anyone's guess.

Perhaps it is time to realize that standardization of business processes across a global enterprise is a pipe dream. While general ledgers may indeed be the same, the idea that markets in Italy are the same as those in Japan, or that the business process that makes sense in a huge developed country is appropriate for a new market in a small country, is surely absurd. Even if you could persuade the business managers in a small, developing country to adopt the processes defined in central office, are these not likely to impose unwieldy constraints on the business? And how likely is it that the central office analyst has thought out a strategy that will work equally well in Beijing as in Baltimore?

Capturing the underlying business transactions through robust ERP systems is a good thing. Using it as a means to stifle innovation and lock businesses in to the processes devised by a software vendor is not. It makes sense to standardize finance systems -- after all, the principles of double entry book-keeping were established by the Venetians. But marketing?

Also, suppose you operate in an environment where you make a lot of acquisitions or divestments (and some large companies make dozens of these a year). Even, suppose, through a Herculean effort you achieve standardized processes within your own company, you can be utterly certain that your shiny new processes will not be the same as those of the companies you have just bought. How will you measure your overall business performance now? Let's see, it takes at least one or two years to re-implement ERP in the acquired company, but we are buying a new company every quarter or so. How long until the nirvana of standardized processes? You do the arithmetic. You will be chasing your tail forever.

Standardization initiatives miss the point. What mostly is required is not the rigid standardization of business processes for its own sake, but the ability to see how the business is performing even though the businesses in different countries are not the same. A modern data warehousing strategy that manages the inevitably different business structures in varying countries will in turn allow "apples and apples" comparisons despite the diversity of systems. For example, you should be able to see the overall profitability of products by channel, customer or geography, even though the underlying data structures in the source ERP systems are different.

As long as you have the architecture and software in place that allows you to analyze data even across differing structures that change, you simply do not need to force all your operating companies into a straightjacket. But don't tell that to the management consultants who are about to make a second fortune by re-implementing the ERP systems they put in the late 1990s; they might not like it.