FSN: Reconciliation software matches expectations

Reconciliation software matches expectations

8th October 2007
The world of reconciliation software does not often make the headlines yet the daily grind of checking balances with suppliers, comparing cash book entries with the Bank and matching inter-company transactions is a pervasive and time consuming activity.

For businesses with very high transaction volumes, i.e. in excess of hundreds of thousands to tens of millions of transactions per day the management of reconciliations can become a major burden. With volumes as high as these any delays in reconciling transactions can quickly lead to unmanageable backlogs, control failures and economic loss. In the face of increasing regulatory scrutiny, stringent auditing standards and Sarbanes Oxley, few companies can afford to slip up.

But it is not simply a matter of compliance. Managing reconciliations more tightly and efficiently makes good business sense. Best practice controls and workflow reduce the risk of error, improve recoverability of old balances, and through appropriate automation reduce the pressure on hard pressed finance functions – liberating more time for value adding activities.

It's a situation very familiar to Trintech plc, a company with a deep heritage in specialised financial applications such as ‘chip and pin,' but in more recent times has applied its knowledge to the knotty problems of governance, reconciliations and compliance (GRC). Its ReconNET product is one of the leading packages in the commercial sector for managing high volume reconciliations, filling the gaping hole left by major ERP packages that fail to provide adequate functionality.

Although vitally important, Trintech's specialised application is not confined to common-or-garden bank reconciliation. Its software is suited to reconciliations in any high volume environment, especially where there is a complex requirement to manage multiple data sources and a need for sophisticated matching across multiple data fields. This takes ReconNET into applications as diverse as fixed asset reconciliation of movements, inventory, inter-company and debtors. But the solution is also finding favour in BPO (Business Process Outsourcing) in Shared Service Centres, where perhaps hundreds of different reconciliations have to be performed concurrently.

Surprisingly, adding value to the reconciliation process is not just about matching reconciling items. Anthony Rafter , a business analyst at Trintech told FSN, “People are impressed with the speed with which we can match thousands of reconciling transactions, but the true value comes in the honing and refinement of the matching process - we call it a ‘sophisticated learning'. Increasing successful matches from 70 through 90 percent or more on millions of transactions translates into massive savings in time and effort, allowing skilled staff to be redeployed to other more productive tasks.”

But the benefits do not stop there. “Resolving unreconciled items can translate into fewer write offs of debtor balances, more accurate management reporting and possibly a reduction in customer complaints. For example, using our software, customers in the telecoms industry have been able to reconcile interconnect charges (transaction charges for using infrastructure) for the very first time with complete confidence.”

So how does the technology work? In broad terms it starts with complete mastery of all inbound data sources. Like a glorified ETL (Extract, Transfer and Load) tool the “Universal File Processor” or UFP allows virtually any data source to be defined with absolute precision. Effectively inbound data can be dissected, manipulated, managed and presented to the target system(s) against which it is being matched. The UFP handles many of the complex relationships that can arise. For example input data may match reconciling data on a “one-to-one” basis but very often several bits of inbound data need to be matched against a single element in the target system, i.e. “many to one” and more complex scenarios such as “many-to-many” are also commonplace.

The key strength of Trintech's solution is that it owns the technology. This means that the specialist has built up a store of knowledge and support around how to manage reconciliations. But the users of the system do not have to be specialists. The way in which data, matching rules and mappings are defined in the system is handled through a straightforward user interface that is well within the grasp of a finance specialist. This means that when the shape of inbound data changes, or rules need to be refined to improve reconciliation success, then the necessary changes can be made by the end user without seeking assistance from IT specialists. The business advantage is compelling because the personnel making the changes really understand the data they are working with and do not have to specify the requirements to an IT department. By the same token the IT department is relieved of supporting yet another application.

Even though the automated matching is very impressive Trintech's Rafter plays down its significance. “So what?” he says. “The value in this process is around improving the success of matching and understanding the exceptions or items that do not match.”

This is why the ReconNET system has a vast armoury of capability for researching why transactions do not match. “Smart Mark” technology suggests ‘near misses' where say four out of five key fields match, but the fifth is out by a few pennies. The user is guided by a traffic lighting system to possible matches that can be displayed side by side on screen for ease of comparison. Close matches are ranked and presented in order of merit and provided the data is present, additional fields not directly involved in matching can be used to help unravel the cause of a mismatch.

The whole aspect of “research” (the term Trintech uses to describe the management of exceptions) is supported by workflow. In other words different classes of exception can be automatically routed via email alerts to other users to resolve. This could be by geography or product type or perhaps by the size (value) of the mismatch. Exceptions requiring investigation appear in users' task lists and subsequent approval of reconciling items and suggested postings can be routed automatically to supervisors for review. (In fact the system can create and post the reconciling entries if required). The whole research system is web-enabled so that all of the functionality is accessible via a web browser which renders the solution capable of being widely distributed at reasonable cost.

Naturally all interventions on the system are auditable, time stamped and marked with the relevant user. A very comprehensive suite of reports provides valuable statistics about the percentage reconciliations and time to reconcile which can be used to guide continuous process improvement. But the system also generates a fully documented reconciliation which evidences the work that has been done and the reconciliations can be passed to other compliance systems such as Trintech's own AssureNET GL.

One of the most important things to grasp about the solution is that it is a reconciliation ‘engine' rather than an application. This means that its very broad capabilities can be applied to any high volume reconciliation process. This is why ReconNET turns up in all sorts of industries from mobile phone operators (O2 and Vodafone), to financial services (Lloyds TSB) and travel (MyTravel, eBookers and TUI).

Reconciliations may not be at the glamorous end of business activity but they underpin the reliability of many key processes on which management depends. Costly and labour intensive to manage, routine reconciliations are ripe for improvement. Reconciliation software is one of those technologies that not only pays for itself in a very short period of time but simultaneously allows management to improve control and compliance. Manual processes are simply no match!

About Us