Maidstone Borough Council Saves £40,000 per Year Using Amazon Textract

The Maidstone Borough Council (MBC) was looking to innovate its document processing to better deliver local services, like waste collection, building control, and tax collection. MBC serves around 70,000 households in Kent, U.K., and its legacy processes for incoming documentation required complicated manual oversight. To increase efficiency and staff productivity, MBC worked to implement an automated request allocation service built on Amazon Web Services (AWS). Developed by Firemind, an AWS Partner, the solution uses Amazon Textract, a machine learning service that automatically extracts text, handwriting, and data from scanned documents, to provide 24/7 online document services for residents and save MBC time and money.

The Maidstone Borough Council (MBC) supports a regional population of 130,000 residents with a variety of core services, all of which involve online applications and forms submitted by users. With 150 processes online, digital contact is the most common form of contact residents use to communicate with MBC. “Depending on the process, 70–90 percent of customer contact comes in through the online channel and our website,” says Paul O’Grady, digital services team leader at MBC. MBC was receiving 300–500 documents a week, and very few submissions were ready to be entered as is into the system. As a result, MBC was inundated with incoming documentation that, despite having arrived digitally, still required hands-on evaluation, processing, and adjustment.

Manual tasks included image rotation, data collection, and PDF merging. In addition, many residents would send in noncompliant image attachments, which needed to be altered before MBC could successfully process them. To reduce the manual intervention needed for incoming documents and emails, MBC engaged AWS Partner Firemind. MBC wanted to work with a local company, and it was impressed by the attitude and flexibility of Firemind. “What Firemind provided was a sounding board and an opportunity to do things differently,” says O’Grady. The proof of concept that Firemind provided to MBC used several AWS services to achieve secure digital documentation with high ease of use and low cost to MBC. Initially, Firemind trialed the solution with 20 documents. Since then, MBC has adopted it for 15 different internal processes.

The first step of the new data processing solution stores the information that residents submitted using digital documentation processing in Amazon Simple Storage Service (Amazon S3), an object storage service offering scalability, data availability, security, and performance. The system can add online submissions to Amazon S3 storage automatically, but MBC can also scan paperwork manually when residents deliver it to MBC offices. The council pulls the documents from Amazon S3 using AWS Lambda, a serverless, event-driven compute service that lets users run code for applications without provisioning or managing servers. By using AWS Lambda, this solution can automatically run code in response to events at virtually any scale to match MBC’s needs. AWS Lambda handles most of the processing, from pulling images and PDFs out of storage to automatically deleting files after 10–15 minutes to comply with the General Data and Protection Regulation in the EU. “We stand by our users’ security,” says Charlie Hudson, managing director at Firemind. “We follow strict security standards, so people can send driving licenses, home location details, and medical records.” Because AWS Lambda is serverless, there was no need for MBC to provision or manage additional servers. Along with the enhanced efficiency of automating document processing, this serverless architecture saves MBC £40,000 a year. The council can reinvest that money to improve other aspects of its services.

The solution also relies on the artificial intelligence and machine learning capabilities of AWS services to analyze documents. MBC uses Amazon Textract to extract text from digital and handwritten documents provided by users. This machine learning service can also identify, understand, and extract data from forms and tables. Using Amazon Textract has significantly reduced the workload: MBC has been able to automate the equivalent of one employee’s work. MBC is also using Amazon Rekognition, which offers pretrained and customizable computer vision capabilities to extract information and insights from images and videos. Together, Amazon Textract and Amazon Rekognition pull relevant data from documents submitted to MBC without the need for manual intervention. This automated process runs 24/7, which further benefits MBC and the residents it serves. “There is no longer a bottleneck in the workflow on Monday mornings,” says O’Grady. “People can use Council services 24/7 while we’re still working nine to five.” The completely automated workflow is processing 55,000–60,000 documents a year on AWS, accounting for 25 percent of all MBC scanning. That frees up employees to focus on helping residents with other important tasks, and the system can scale further if needed.

The solution that Firemind provided MBC is a managed service, and Firemind has access to the account so that it can quickly respond if any issues arise. As a result, the process is a turnkey solution for MBC, with no staff upskilling required to deploy or run the service. “It’s a simple solution with a great impact,” says Hudson. “When you show a company what it is capable of on the cloud, adopting AWS services adds massive value.” Firemind built the solution using AWS services and worked with MBC to make its components function smoothly as part of the council’s workflow. This tool is simple to implement for many similar organizations, making it possible to reduce manual work using the automation capabilities of AWS with little overhead and fast deployment.

Credit: Firemind

MBC is looking to innovate further by expanding its use of AWS in the future. “With the workflow that Firemind built, we could fully automate our document process,” says O’Grady. “The experience was positive, and we’re planning future projects using AWS.” MBC is currently working with Firemind on a proof of concept for hosting its websites and microsites on AWS. The ease of implementation of its automated documentation solution and the increased productivity that resulted has encouraged MBC to pursue more agility and flexibility by using AWS resources in future developments.