In the midst of the continuing revelations about suspect accounting practices, inflated earnings and pro forma obfuscation, my colleagues at CGE&Y recently published the results of a May 2002 survey of 265 CFOs from mostly large companieseight out of 10 had more than $1 billion in revenueson their thoughts about the current and future state of the finance function. I'm not going to repeat much of their findings except to point out that a whopping 63 percent of CFOs said they are constrained in their ability to do their work and to transform the finance function because of inadequate information systems.
Excuse me? This is, after all, the second half of 2002. Post-Y2K (didn't we fix everything in finance for them back then?). Mostly post-ERP deployment (didn't we automate everything in finance?). Post-data warehousing (didn't they hyper-analyze everything six ways to Sunday with all those OLAP tools we gave them?). Post-Web-based reporting (didn't we make the data available to everyone with a browser on their PC/PDA/mobile phone?). And they still don't have the systems they need to get the job done? What the heck have we been doing for the past decade?
Okay, I've calmed down. Assuming the survey accurately reports CFO perceptionswe've got to trust somebody's numbers, after allwhat's going on here? Let's do a little root-cause analysis.
First off, the CFOs specifically reported that they had "inadequate . . . budgeting, forecasting and decision support systems." Many of them have trouble when they have to consolidate the financial reporting of the results of subsidiaries that operate under foreign jurisdictions and don't use U.S. generally accepted accounting principles. Clearly, there's a lack of international standards for financial accounting, and that's certainly not the fault of corporate information systems. Quite a few are wrestling with the integration of recently acquired businesses, with restructuring issues, and with new value-chain models and the complex accounting rules they bring (vendor-managed inventory, for example). But they can't blame us if they keep messing things up by adding incompatibilities into the mix, can they? And many CFOs struggle to understand just what their information systems should be doing for them. They can't blame us if they don't know what their own systems should be doing, can they? So this is mainly a people and process problem, and the CIO is off the hook until the finance folks get it figured out, right?
Well, maybe not. Hidden in there, amongst all the process and people issues, are a few items that live a lot closer to home in the IT department. Take the issue of metadata (I can already hear the groans). In too many cases, consolidation of subsidiary financials (even of operating units within the U.S.) is severely hampered by a lack of standard definitions for things in the chart of accounts. This shouldn't be too hard to clear up, should it? Except that we have been trying for two decades and we still can't do a decent job of metadata management. We know that the creation of rigid, monolithic enterprise data models and the centralized management of data architectures don't work very well. But we really need to get these data issues cleaned up if our technology investments are to pay off. We can only blame the various national and international accounting standards boards for so longsooner or later they are going to agree on one set of rules, and then we'll really be exposed.
Data quality isn't the only problem, apparently. There are a lot of broken business processes out there as welland where processes aren't broken, they're often not very well connected to the technologies that automate them. To listen to the CFOs, it's as if all that process reengineering, Y2K remediation and ERP deployment went on right alongside them but never got connected to the work they actually do.
It gets worse (you knew it would). CFOs report that too much staff time in finance (nearly 40 percent) is spent on transaction processing. Didn't we automate all that a couple of years ago when we let everyone place orders and track order status online? Turns out all that inconsistent data and all those broken or disconnected processes cause lots of exceptions that need to be dealt with and lots of reconciliation that needs to be done. And that takes lots of peoplealbeit people increasingly sitting in an outsourcer's facility in Ireland or India or the Philippines. It all costs money, though, no matter who does it.
Finally, CFOs think there is too much automation out there, in the sense that there are too many different systems and products installed for doing essentially the same thing. Too much diverse automation reduces flexibility (because you can't easily move finance people from one part of the company to another without retraining them on different technology) and increases complexity (because all these systems have to be hooked together somehow). This in turn leads to increased costs (which also hampers the efforts of CFOs to transform their systemsseems they have budget problems just like the rest of us). CFOs ask, not unreasonably, why the business needs so many transaction, database and reporting tools. After all, how many different general ledger, accounts payable, accounts receivable and asset management systems are needed to get the job done? And every extra set of technologies is another excuse to have bad or inconsistent metadata, disconnected processes and confused people.
Good lord, how did we get into this mess? And finance and accounting is easy compared with lots of other things we're supposed to have automated. Did we foul up just as badly everywhere else? How long before the CFO starts asking that question?
Okay, I'm calm again.
If your CFO is really thinking this way, what can you do to make his or her life easier? For starters, some (re)centralization strategies, some simplification proposals and some standardization efforts will probably be looked on favorably. Conversely, what are you or your business users planning that will, in the eyes of your CFO, make things worse? I don't think I'd want to be making too many suggestions in that direction right now.
Maybe this is the time to try and get back some control over all those wayward business units that keep buying new technologies because they are "best of breed" even though they also are "most expensive to own" once you factor in the integration and support costs your users never seem to factor into their business cases. Maybe it's time to get out that list you keep of all the recalcitrant business unit executives who demand their own customer numbering scheme or their own SKUs and product codes or their own HR rules. Maybe this time we can get that data harmonization and reconciliation project to really work.
And let's not forget all those users who can't or won't use a standardized processbecause they never had to in the past and they don't want to change now. Maybe we can use CFOs' little problems to get them to fall into line as well. Then, with all the money we will have saved CFOs in running the finance function, maybe they will let us start spending some real money again.
Imagine how slick a reengineered finance function could be if we fixed all of these information systems issues for the CFO. Real-time financial monitoring with automatic alerts and notifications. "Close the books on demand" capability. Virtually error-free, frictionless transaction processing. Business-unit and product-line performance simulation for closed-loop controls. Automatically updated performance scorecards. Continuous budgeting. Online, up-to-date reporting of results, both internally and externally. Low operating costs and high business impact. All the above and more are what CFOs want from the finance and accounting functions in the future.
And maybe, after we are all done, CFOs will have some numbers they (and everyone else) can trust.
John Parkinson is chief technologist for the Americas at Cap Gemini Ernst & Young. Please send comments on this story to firstname.lastname@example.org.