Efficient and accurate reconciliations are critical for ensuring data consistency, enhancing client satisfaction, and reducing operational risks, primarily where multiple copies of the same trade data are held across financial institutions and their customers.

Within the financial service industry, large-scale reconciliations platforms competently undertake substantial swathes of the reconciliation process, such as positions, transactions, and cash, and bring financial institutions to a favourable position. Favourable but not flawless.

Outside of these enterprise-wide reconciliations we are witnessing a proliferation of complex, non-standard, unstructured, and EUC-styled reconciliations. Together they form a growing percentage of the overall reconciliation processes happening within financial institutions. Moreover, they are laden with risks and take up excessive resource and time to resolve.

Typically, ‘offline’ and dependent on spreadsheets, macros, and heavy IT team involvement, these awkward, peripheral variants tend not to make it into larger reconciliation tools for several reasons. 

• The tools can’t cope with unstructured documents: Rarely can traditional reconciliations tools extract semi-structured or unstructured data, such as that compiled within emails and PDF documents.
• ‘Non-standard’ reconciliations can’t be accommodated: The likes of Security Master reconciliations, regulatory reconciliations for Securities Financing Transactions Regulation (SFTR), or invoice reconciliation are often forced into a data structure that does not represent the reconciliation type (a common example encountered is forcing an invoice reconciliation into a cash reconciliation).
• Rogue data can’t be cleaned: Upfront preparation or transformation of data that needs cleaning to avoid false breaks is mostly unavailable.
• Extra IT development is needed: Setting up a reconciliation within a traditional tool is often time and labor-intensive, with IT teams taken off tasks to configure the required reconciliation rules. 

Problems begin to mount

The complexities inherent to the ‘offline’ reconciliation space are growing to such an extent that financial institutions are increasingly pressed into partnering with multiple reconciliation vendors.

Inefficient, it is an approach that also becomes expensive as companies become mired in various licence fees for inadequate systems. Additionally, they find themselves locked into platforms that take too long to onboard new reconciliations or cannot handle their complexity and volume.

Against this convoluted and costly backdrop, financial institutions must either rectify failed reconciliations manually, causing delays, losses, and risk exposure, or wait for technology teams to become available to develop new, complex reconciliations systems - a process that can take months, if not years.

Ultimately, control is lost, auditability collapses, and key person dependencies skyrocket.

Automating the edges

Time-consuming, expensive, and risk-laden though they are, as 'offline' reconciliations exist on the peripheries of the broader reconciliation process, they have come to be accepted as an unavoidable burden — the industry's own immovable rain cloud.

However, in recent years, solutions have emerged with functionality focusing specifically on the data within this most neglected and troublesome category of reconciliations.

Data is the industry's biggest challenge; how it can be aggregated, validated, curated and enriched from multiple sources, particularly PDFs. Without complete and readable data, no reconciliation can be completed effectively. Indeed, ensuring that such data is available for the reconciliation to progress is typically more challenging than performing the reconciliation itself. 

Using the latest Optical Character Recognition (OCR) and Natural Language Processing (NLP) techniques, modern platforms ingest reams of structured, semi-structured, and unstructured data from multiple sources regardless of the document types. These data centric platforms automatically convert sources into data that is aggregated, validated, and accurate from the start.

Ending the Remediation Conundrum

To 'automate' a process is to infer that it becomes faster and easier to manage and that this alone is reason enough for its implementation.

True that speed and simplicity are two significant benefits of automation, reconciliations – especially high-value ones – are too sensitive to simply feed into a machine and hope for the best.

Among the industry’s more frustrating challenges is the need to remediate reconciliations as it generally means that something within the data was wrong. Where remediations begin to mount or take too long to resolve, the financial and reputational costs usually escalate. Even if machine learning is deployed post- event to remediate a reconciliation, it amounts to a delay.

A crucial aspect of next-generation platforms is their ability to automatically interrogate, standardize, and validate data before the reconciliation process begins. As such, any remediations become a true exception and provide a clear signpost to a potential risk event.

Owning the Process

Despite the reconciliation process being dependent on data that operations teams know better than anyone else within a company, the complexity of legacy systems means much of the work must be undertaken by IT departments instead.

Onboarding reconciliations based on unstructured data into outmoded solutions can take months (even years), thereby removing IT teams from other, more value-adding tasks and pulling operations teams away from a process that should be theirs to own.

With new technology allowing operations teams to easily prepare offline reconciliations and then configure the rules needed for their proper execution, the process is accelerated, the chance of error is reduced, and IT teams can return their focus to growth and innovation.

Regulatory Pressures are Building.

The business incentives for bringing automation to the offline reconciliations space are compelling.

Apart from the process being cut from taking months to taking only days or hours, risk of error is slashed, key person dependency is reduced, and time to concentrate on more mission-critical tasks is freed.

That enhanced compliance is also provided becomes a more urgent advantage when considering the incoming pressure that T+1 will apply to operations teams.

The transition from T+2 to T+1 halves processing times and is already triggering widespread change among market participants across infrastructure, technology, and behaviour. It will become essential for financial institutions to execute faster reconciliations to maintain compliance, an objective that cannot be achieved when offline reconciliations reside within Excel spreadsheets.  

In a T+1 world, certain tasks, actions, and processes face significant disruption. Possessing the tools needed to manage this disruption - one that covers all areas of a company’s operations - will allow for the development of a more comprehensive strategy for dealing with all T+1 contingencies.

Pressure, however, comes from beyond the anticipated rigours of T+1. In a letter to bank and building society CEOs on his thematic findings on the reliability of regulatory reporting, David Bailey, Executive Director for UK Deposit Takers Supervision at the Bank of England, covered reconciliations explicitly.

Aside from sharing his concerns with the reliance on spreadsheets, he stated: “Reconciliations are an essential element of generating reliable regulatory returns, and we observed unsatisfactory reconciliation disciplines across a number of firms. We expect firms to have a formal and comprehensive process reconciling regulatory flows to appropriate records, including the general ledger, for every submission cycle.”

He concluded, saying: “We expect firms’ remediation plans to be strategic, appropriately resourced, and address the root causes of issues.”

Taking the Edge off the Reconciliations Process

As technology is advancing at a rate similar to regulatory evolution, there is no longer the need for financial institutions to find themselves trapped in a web of providers, legacy systems, and manual work.

By bringing next-generation automation to the edges of the reconciliation process, working practices are transformed allowing financial institutions to deliver a better client experience while reducing operational risks.