Driving Financial Services Customer Success with a Cloud-based, Smart Metadata Hub

One financial institution that provides personal and commercial banking, wealth management, and financial market services to their clients, leveraged Semaphore Cloud’s SaaS platform to solve their information challenges.

Semaphore Cloud provides a flexible solution and supports their ability to geographically segregate enterprise data, which allows the organization to meet search and compliance requirements. The environment is cost effective as it eliminates expenses associated with hardware acquisition, provisioning, and maintenance. And it reduces the maintenance hassles of software installation and support of applications on company computers or in data centers.

As part of their initiative to migrate the company intranet to Office 365/SP Online, they identified a need to evaluate the efficacy of manual content classification. Internal users performed manual tagging of information assets yet the results were not consistent. When documents spanned multiple pages, they were broken into smaller documents, which increased the corpus for manual tagging and user workload. The result – an increase in tagging errors.

To be successful, they needed a semantic platform that would support auto-classification in multiple languages – French and English - enhance and enrich collaboration and knowledge management environments, as well as provide augmented model harmonization, management, and governance capabilities.

They began by using Semaphore Knowledge Model Management (KMM) to create a multi-language (French & English) enterprise model to support a broad range of use cases and reflects the policy documents, procedures, and end user content throughout the organization. They worked with multiple stakeholders i.e. Human Resources, IT, Project Management, and subject matter experts, to enrich the model with synonyms and alternative labels that reflect the content to improve model quality. The model was further enhanced with additional vocabularies and relationships between vocabularies were identified.

The model was published and combined with AI, natural language processing, and machine learning strategies to perform bulk classification of site collections as well as on-going, auto-classification of content stored in SharePoint Online. This robust classification strategy incorporates organization data, conversations happening throughout the enterprise, and end-user content stored in Microsoft repositories.

Semaphore’s ability to quickly auto-classify content and apply precise, complete and consistent metadata, results in transparent and accurate results. Semaphore’s Classification Review (CRT) and Classification Analysis Tools (CAT) document and analyze classification variations and address current classification challenges. The ability to classify in automated as well as assisted mode, allows users to overwrite tags based on a comparative list, which greatly increases user productivity.