N-way reconciliation: Deep Dive

For years, n-way or multi-way reconciliation has been a key requirement among banks and investment managers looking to accelerate the process and reduce errors. But do you know if you are doing true n-way reconciliation? What do n-way or multi-way actually mean, anyway?

What is true n-way reconciliation?

Some reconciliation software platforms use two-way or three-way reconciliation and call them ‘multi-way.’ Neither are the same as n-way reconciliation nor do they satisfy a true n-way requirement. 

Multiple two-way reconciliations are simply multiple source comparisons that are viewed separately. This approach consumes more processing power, increases onboarding time and reduces operational efficiency, ultimately causing delays and errors.

Many reconciliation solutions offer three-way reconciliations. Or they concatenate multiple data sources into a single file, requiring multiple, arduous steps which involve merging different ledgers into a single file, matching them against a single statement file, then matching that file against the other side.

In contrast, true n-way reconciliation compares multiple unique data sets within the same view, allowing you to compare large volumes of data quickly and seamlessly.

The gravity of pseudo “multi-way” reconciliation.

Adding data sources to a pseudo multi-way requirement further complicates the process because many systems require you to create one-to-one reconciliations for each of the sources involved. 

Say your reconciliation platform has three data sources and you want to perform one-to-one comparisons between each source. You would need to perform three comparisons to reconcile the data as follows:

• Comparison 1: Compare data from Source A with data from Source B
• Comparison 2: Compare data from Source A with data from Source C
• Comparison 3: Compare data from Source B with data from Source C

However, anything more than three data sources will require significantly more comparisons. How many, exactly? Well, the formula for calculating the number of comparisons required is n(n-1)/2, where n is the number of data sources. With five data sources, that’s 10 one-to-one comparisons (5(5-1)/2 = 10) you would need to do, as follows:

• Comparison 1: Compare data from Source A with data from Source B
• Comparison 2: Compare data from Source A with data from Source C
• Comparison 3: Compare data from Source A with data from Source D
• Comparison 4: Compare data from Source A with data from Source E
• Comparison 5: Compare data from Source B with data from Source C
• Comparison 6: Compare data from Source B with data from Source D
• Comparison 7: Compare data from Source B with data from Source E
• Comparison 8: Compare data from Source C with data from Source D
• Comparison 9: Compare data from Source C with data from Source E
• Comparison 10: Compare data from Source D with data from Source E

Supporting this scenario is a nightmare for your IT department and your end users in terms of configuration, maintenance and lack of efficiency.

It’s also an example of why Gresham’s Control is vastly superior to solutions that do not support true n-way reconciliation. A solution that provides native support for true n-way reconciliation scenarios (i.e., ABOR, IBOR, banks, etc.) should simplify the creation and management of reconciliation processes that include three or more sources.

The bottom line

It’s important to know what your n-way recs will look like and how they will be built to avoid unpleasant – and potentially costly – surprises. Your system should be data agnostic and easily scalable, and enable n-way reconciliations to be built easily in one single solution. Only then can your organisation alleviate the strain of pseudo multi-way reconciliation scenarios.