FSN: Financial data quality – the lynchpin of dependable reporting

Financial data quality – the lynchpin of dependable reporting

17th March 2008
Data quality is the equivalent of haemorrhoids to the medical profession! It causes embarrassment, can have painful consequences and takes time to cure. Yet without robust data quality the integrity of financial reporting processes is completely undermined. In fact as reporting requirements broaden, become more complex to embed and are demanded in ever quicker timescales the risk of reputational and financial damage grows exponentially. Gary Simon , FSN's managing editor looks at the problem and approaches to solving it.

The heavy burden of new regulation in the last five years has forced almost every company around the world to collect and maintain more information. Just last week, Deloitte, the accounting firm reported that half yearly financial reports in the UK had grown 27% in just a year! There is every reason to suspect that other geographies have suffered from exactly the same phenomenon.

What is often lost within this rapidly growing trend is the impact of broadening information requirements on the number of data sources that have to be managed. Non –financial reporting, such as Corporate Social Reporting, Environmental Reporting and Employee disclosures, together with relevant KPIs, have to be gathered form a wide range of exotic data sources that span a number of different functional areas. The days of the ‘fat ledger' where every bit of information could be collected through the general ledger system have long gone. However, recent mergers between ERP vendors and performance vendors offer some prospect of returning some semblance of order to a fairly wild data landscape.

In the past, ERP systems and performance management systems have occupied different domains – both logically and physically. From a logical point of view one environment deals with transactions whereas the other works with summary balances. Physically, the applications rarely share the same computing environment. But if you let your imagination run wild for a minute it is possible to foresee that with ERP and performance management systems in the same supplier stable, for example, Oracle, SAP and Microsoft, the potential for sharing information could become simpler. It is a huge ‘Ask' and may be several years away but at the very least it would be extremely convenient if metadata (structural information about, say accounts, periods, currencies and so on) was common across the whole environment.

At the moment such a suggestion appears quite farfetched, mainly because most performance management vendors struggle to harmonise metadata in their own world let alone in the ERP space as well. Ask yourself whether a cost centre has exactly the same definition in your budgeting, planning, forecasting, consolidation, scorecarding and dashboarding applications. Then consider ERP modules such as sales order processing, human resources, fixed assets, production and payroll and you begin to get a sense of the scale of the problem. Yet it shouldn't be beyond the wit of man to do this and the recent mergers provide an ideal springboard for this to happen.

Back in the real world how do companies control metadata? The answer is not very well? For example, if you changed a chart of account line in a subsidiary in China would your group reporting applications ‘trap' the change at the earliest opportunity or simply propagate the problem up the Reporting Supply Chain making it progressively more difficult to correct the error? In most organisations any information related to the changed account details would simply get rejected and lay unattended until someone spotted the error. But there are more positive ways of dealing with this type of problem.

ETL (Extract, Transform and Load) tools are the IT department's preferred way of managing the flow of data from source to target systems, such as a subsidiary's ledger to a group reporting pack. But these are rarely finance-friendly. In the main they are designed for technicians and use the language of IT. What is needed is ETL type functionality but packaged up with an interface that buffers the finance function from the complexity.

The tools are starting to appear. Before Hyperion merged with Oracle the company had acquired a company called Upstream which provided exactly the sort of tool needed. Now a mainstream part of the Oracle/Hyperion solution set the application gives the finance department visibility into the reporting supply chain from data capture in subsidiaries right through to consolidation. In fact the flow of information is bi-directional so that enquiring on a final consolidated number allows the user to drill back down through every interface and right down to the originating transaction. Not only is the audit trail comprehensive but if the accounts either side of the interface change, for example, because Group or a subsidiary have changed an account, then the change is trapped and can be repaired ‘on the fly' with all of the interface owners involved. Powerful technology such as this provides a double whammy – firstly it improves data quality and secondly it accelerates information flows.

With data volumes growing all of the time data quality has to take centre stage. Investing in metadata management may not sound appealing but it is worth every penny.

About Us