CASE STUDY | HEALTHCARE
Manchester University NHS Foundation Trust Fully Automates Invoice Processing and Reduces Payment Risk
Manchester University NHS Foundation Trust (MFT) is one of the largest acute trusts in the UK, operating 10 hospitals and employing over 20,000 staff. With smooth, reliable communication between ABBYY IDP and its ERP system, MFT has reduced invoice processing time and freed its staff to focus on other business-critical areas.
Edd Berry, Director of Finance Innovation, Manchester University NHS Foundation Trust
Challenge
MFT processes around 275,000 invoices a year from approximately 6,000 live suppliers, with thousands of different types of invoices. It planned to further automate invoicing by feeding information from its existing IDP solution to its ERP system, enabled by robotic process automation (RPA). However, manual effort was still required between the systems due to IDP inaccuracies and inconsistencies that could not be processed by the RPA tool.
MFT needed to find an IDP solution that could:
- Automatically locate and extract relevant invoice information with a much higher degree of accuracy
- Pass data reliably to RPA so it can successfully align with POs in MFT’s financial system
- Minimize coding requirements for data manipulation
- Reduce invoice validation and training time
Solution
ABBYY IDP identifies relevant invoice data, even in the most complex documents, with the highest accuracy. The RPA tool can now successfully pass it to the ERP system.
ABBYY IDP allows MFT to apply advanced functions across an entire set of invoices instead of repeating rules on each.
Staff can now build a skill with low—even no— code, leveraging pre-built extraction models and standardized business rules. ABBYY IDP learns from manual document validation, so time spent on training and validating invoices is continually reduced.
The consistent and reliable results achieved with invoice automation has given MFT confidence about extending this automation to processes beyond invoices.
Value
- Solution successfully rolled out in weeks, not months
- Reduction in processing time, payment risks, and monthly reconciliations
- Staff free to focus on other business critical areas, rather than coding
- Continuous machine learning reduces post-processing and data validation time
Edd Berry, Director of Finance Innovation, Manchester University NHS Foundation Trust
Challenge
MFT processes around 275,000 invoices a year from approximately 6,000 live suppliers, with thousands of different types of invoices. It planned to further automate invoicing by feeding information from its existing IDP solution to its ERP system, enabled by robotic process automation (RPA). However, manual effort was still required between the systems due to IDP inaccuracies and inconsistencies that could not be processed by the RPA tool.
MFT needed to find an IDP solution that could:
- Automatically locate and extract relevant invoice information with a much higher degree of accuracy
- Pass data reliably to RPA so it can successfully align with POs in MFT’s financial system
- Minimize coding requirements for data manipulation
- Reduce invoice validation and training time
Solution
ABBYY IDP identifies relevant invoice data, even in the most complex documents, with the highest accuracy. The RPA tool can now successfully pass it to the ERP system.
ABBYY IDP allows MFT to apply advanced functions across an entire set of invoices instead of repeating rules on each.
Staff can now build a skill with low—even no— code, leveraging pre-built extraction models and standardized business rules. ABBYY IDP learns from manual document validation, so time spent on training and validating invoices is continually reduced.
The consistent and reliable results achieved with invoice automation has given MFT confidence about extending this automation to processes beyond invoices.
Value
- Solution successfully rolled out in weeks, not months
- Reduction in processing time, payment risks, and monthly reconciliations
- Staff free to focus on other business critical areas, rather than coding
- Continuous machine learning reduces post-processing and data validation time