HM Land Registry Cuts Document Review Time in Half Using Amazon Textract

HM Land Registry (HMLR) reduced document review times by 50 percent after working alongside AWS Advanced Partner Kainos Software Limited (Kainos) to build and implement an automated comparison tool on AWS. The government agency’s caseworkers need to review complex legal documents manually, which was time consuming, so HMLR and Kainos built a solution powered by Amazon Textract and Amazon Comprehend that automatically compares documents and flags discrepancies for review. Now, caseworkers no longer need to review thousands of documents per week, and the agency can approve property transfers faster.

HMLR safeguards more than 25 million property titles that show evidence of land ownership. When a new building development is planned, HMLR caseworkers must compare the transfer deed submitted for registration with precedent transfer deeds to confirm that there are no significant discrepancies. Document comparison is manual and repetitive, and differences between documents are often minor, making the process error prone. Additionally, the documents’ legal language is often very complex, which further complicates the task.

The agency realized that artificial intelligence could be used to automate this process. It turned to AWS Advanced Partner Kainos Software Limited (Kainos), a digital solutions provider based in Northern Ireland. After exploring ways to automate document comparison using optical character recognition, Kainos chose to develop an application on AWS. “We realized that Amazon Textract would be the most straightforward to configure,” says Piers Campbell, head of technology, data, and artificial intelligence at Kainos. “This service delivered the best results out of the potential services we identified.”

The web application uses Amazon Textract and Amazon Comprehend, a natural-language processing service that uses ML to uncover information in unstructured data. HMLR caseworkers simply upload transfer deeds as PDFs. The solution then uses a range of ML and artificial intelligence techniques to automatically identify discrepancies between documents.

Using the natural-language processing capabilities of Amazon Comprehend, the application can extract meaning from complex legal language, identify minor differences, and flag issues for caseworkers to review. By offloading manual work from caseworkers, who previously needed to compare thousands of documents each week, HMLR has doubled its document review speed and can approve property transfers faster. This solution also reduces the risk of indemnity claims: it flags discrepancies early in the application process, prompting caseworkers to resolve issues before they evolve into legal disputes.

By adopting the AWS-powered comparison tool, HMLR has reduced document review times and increased staff productivity. In the future, HMLR and Kainos plan to use caseworker feedback and data collected by the solution to further refine and improve the ML models. “This project has successfully shown how we can use artificial intelligence and machine learning and, more importantly, how we can strengthen that capability as we progress through our digital transformation,” says Nick Davies, senior product manager for internal digital services at HMLR. “Using AWS, we can improve our processes and enhance how our employees work.”