Archival Appraisal for an Environment Agency

Through enlisting Castlepoint to manage their archival appraisal, a division of a government department responsible for environment, energy, and climate change was able to drastically expedite the classification process as well as identify significant quantities of records which can be compliantly disposed of without diminishing the value of the data set or affecting operations.

The Challenge

The Division is a part of the Department responsible for the environment, energy, and climate change. The Program it manages is a collaborative partnership across government and more than 150 national and international research institutions, focused on researching the world’s climate and the effects of climate change. The Division manages significant amounts of structured and unstructured data, from many sources, much of which requires long term archival preservation.

The Solution

The agency implemented Castlepoint on-premises across their archival file share to automate the appraisal process. To facilitate deployment, Castlepoint extracted over 100 million key phrases from the content, utilising these phrases alongside the limited available metadata to classify the divisions records for either disposal or preservation.

The platform delivered:

  • Legacy data appraisal for migration
  • Automated record sentencing with AI
  • Data minimisation of personal information
  • Automated identification of high-risk data
  • Agentless management of multiple systems
  • Generative AI query and source data audit

The Outcome

Castlepoint identified approximately 25% of the records as eligible for compliant disposal under law, without diminishing the value of the dataset or affecting operations.

This outcome significantly optimised the agency's data estate, ensuring that high-risk and low-value content was removed to enhance privacy and security. The remaining high-value research data was accurately classified for long-term preservation and migration, improving the quality of future outputs and preparing the dataset for future Generative AI deployment.

Key results included:

  • 27,000 records managed with Artificial Intelligence
  • 135.3 million key phrases extracted from over 6.2 million items
  • 25% of records identified for disposal
"Scientific and research organisations like universities, institutes, and environmental agencies have a need to capture and retain enormous data sets. But these can hide PII and other sensitive information that needs to be either destroyed or anonymised as soon as possible to protect the privacy of researchers and other stakeholders. As well as this, disposing of redundant or low-value source data also improves the quality of future outputs, especially as GenAI starts to be deployed to resurface old content."
Rachael Greaves, Castlepoint CEO

Our team are experts too. We love to help.

Get in touchJoin Our team