Customer Stories
.png)
_.png)
.png)
.png)
Tier 1 bank enhances trade settlement reconciliation
Client
The global operations cross-business trade settlement function of a tier 1 investment bank with over $3 trillion assets under management.
Reconciliation type
Daily FX and derivatives trade settlement cashflows vs bank statements across all traded currencies and more than 1200 bank accounts.
Current customer solution
In-house built ML based matching system.
Key Statistics
10
FTE's freed up
At > 1 minute per manual match the client saved 70 user hours per day (10 FTE's) resulting in completion of the daily reconciliation hours earlier than previously.
70
Hours saved daily
10 FTE's freed up per day by reducing the manual match workload by an average of 4000 bank transactions per day (66% reduction).
Customer Challenges
The customer matching system, with user-supplied tuning, delivered superior match rates (between 80 and 90%) compared to their previous rules-based matching system. Even with its improved effectiveness, out of 60,000 bank transactions the reconciliation team of more than 20 people were still manually matching between 5000-10,000 bank transactions daily.
The operations team described this hours-long matching process as a 'nightmare' due to the high volume of manual interventions required and the tight deadlines for the completion of the daily reconciliation. This was especially acute on certain peak volume days and quarterly financial reporting periods.
As well as manual matching workload the users had to also identify and correct around 200-300 system mismatched bank transactions per day (the existing in-house system having a systemic mismatch rate of 0.3%). These mismatches were in-part due to the uncalibrated confidence values assigned by the in-house ML matching system.
Why did they choose Matchimus
The customer wanted to keep their existing processes in place and on-premises. They first tried using a leading auto-ML system to create a machine learning mode to improve the matching, but did not obtain the results they needed. The client chose Matchimus as an on-premise ML solution that would give them the optimal match rates buy adding it to their current solution without requiring the data migration, user education and process changes associated with a full system replacement.
Solution Implementation
-
Installation: Matchimus was installed onto the client’s existing QA and production machines.
-
Automatic Training: A single customer extract comprising one week's worth of production data was extracted. Matchimus utilized this data to train its algorithms (no preprocessing or data cleaning was required), leveraging its purpose-built machine learning algorithms to understand and adapt to the specific patterns and structure on the bank’s reconciliation data.
-
Integration with Existing Workflow: Crucially, the existing reconciliation workflow was not changed. The results from the additional matches identified by Matchimus were seamlessly fed back into the bank’s primary reconciliation system. This ensured continuity and stability in their reconciliation process without requiring major workflow changes.
-
Model Risk management and Security approval: Matchimus met the Tier1 rigorous and in-depth Model Risk management and security approvable with its comprehensive monitoring reports covering model performance, drift, feature sensitivity
Results
The deployment of Matchimus yielded the following improvements to the customer’s reconciliation process:
-
Reduction in Manual Matching workload: Matchimus boosted the average matching rates from 88% to 95% reducing the manual match workload by an average of 4000 bank transactions per day (66% reduction). At > 1 minute per manual match the client saved 70 user hours per day (10 FTEs), resulting in completion of the daily reconciliation hours earlier than previously.
-
Improved Match Accuracy: Matchimus achieved a match accuracy of 99.99%, reducing mismatches from 200 to 6 per day, a 95% reduction, enhancing overall reconciliation accuracy and reliability and reducing the team’s mismatch correction workload.
-
Team morale improved due to the reduction in mundane tasks at the start of their workday.
Our Clients Say
"The implementation of Matchimus is a "no-brainer" and we're seeing great results in our other Straight-Through Processing (STP) implementations. I wish all our implementations had this guaranteed ROI"
Tier 1 bank, Senior Strategist
Pricing plans to fit your needs
Along with annual license models we also offer ‘pay as you save’ pricing plan options, so you avoid the large upfront capital costs usually associated with software installations and guarantee your ROI.
Module packages range from low-cost micro instances all the way up to enterprise level with source code license options, so whatever your needs we have a package to suit you.
Our transaction-based pricing option is a small fraction of the cost of your manual effort by so your ROI kicks in from day 1, and volume discounts mean the higher your manual matching volumes the lower our per transaction fee and even bigger your savings.
Get a Demo of Operartis
01
Want to get started? Fill in the form to start the conversation about how Operartis can help your business.
02
Want to see proof of value? We'll measure the match rate improvement on your chosen reconciliation data and provide an automation report detailing your ROI before you commit to purchase.
:+1-212-810-0264