About the Project
Following previous engagements focused on assessment of foundational taxonomy and metadata capabilities, the client, a Fortune 500 publisher of business and finance software, engaged Factor to develop requirements for consumption of centrally managed enterprise taxonomies by its internal intranet platform. This project was one of a series of projects with the goal of modernizing and streamlining the intranet’s taxonomy and metadata resources and bringing it into alignment with the enterprise taxonomy environment.
Challenge
The client’s intranet is an important centralized hub for employee information. It provides access to self-help resources from a number of internal sources, including IT assistance, human resources, finance operations, global business services, purchasing, and procurement.
However, the business systems managing this information were not integrated with the central taxonomy service, which meant that vocabularies, and as a result the tags applied to content, were not harmonized with metadata and terminology used in the rest of the organization. The result was that employees searching the intranet for the information they needed could never be sure that they were getting all the relevant results, as a document might use a different word for the same concept. Without some way to unify those equivalent words, a basic search will miss relevant content.
Specific challenges included:
The client was tasked with improving employee access to these self-help resources with the overall goal of increasing employee productivity. We sought to identify requirements to provide fit-for-purpose taxonomies. This would address the primary use cases: supporting content management workflows and automation for intranet stakeholders and information discovery by employees throughout the company.
Specific challenges included:
- The intranet content was aggregated from multiple sources, including business systems like LumApps, ServiceNow, WordPress, Confluence, and GitHub, each with its own tags, content, and capabilities.
- Taxonomies across these systems were siloed, had developed independently, and were not aligned with the enterprise taxonomy program.
- The complete enterprise taxonomy set was not needed for representing intranet content and contained many terms relevant only to other use cases. For example, the primary use cases for the intranet were discovery and navigation, and these capabilities could be supported with specific subsets of relevant terms from the enterprise taxonomies.
- The client was tasked with improving employee access to these self-help resources with the overall goal of increasing employee productivity. We sought to identify requirements to provide fit-for-purpose taxonomies. This would address the primary use cases: supporting content management workflows and automation for intranet stakeholders and information discovery by employees throughout the company.
Factor’s Approach
Gathering requirements for technical projects involves understanding the systems, people, and processes at play. For each of the relevant systems we sought to discover:
- System name and function.
- Who is the responsible point of contact, i.e. the system owner or administrator?
- Who are the system users and what are their roles? In particular, we want to understand how they access and use the taxonomy and for which business needs.
- Which taxonomies are used by the assessed system? What are the use cases: for example, tagging, search support, workflow or lifecycle management? What are each system’s capabilities and limitations with regards to taxonomy modeling and management?
- Taxonomy ingestion process: How are taxonomies currently consumed? Possibilities include API, webhooks, direct connectors, file import, and manual entry. Is this process documented and governed? How often are taxonomies changed or updated?
- Process automation: Is taxonomy ingestion automated? Is automation possible? What are the supported formats for automated ingestion? What are the barriers – technical, process, and resources – to automation?
Collecting and consolidating this information allowed us to develop a coherent roadmap to automation of taxonomy ingestion, which would harmonize the intranet systems with standard enterprise vocabularies.
What We Learned
- Each content authoring and management system had its own set of taxonomies for applying metadata to content. None were consuming updated vocabularies from the central taxonomy service.
- Each CMS was managed separately, with different vocabulary requirements and integration issues.
- In at least one case, an integration to automatically consume enterprise taxonomies was attempted, but abandoned after an initial failure. The failure was due to the presence of multiple instances of the content management system in the client’s environment, each with its own unique configuration. API taxonomy integration that worked for one system failed for another.
- The enterprise taxonomy management platform has robust capabilities for reuse of concepts across projects and supports use of multiple labels for concepts. These are important capabilities to provide fit-for-purpose taxonomy subsets for specific use cases.
- The enterprise taxonomy team, including resources from IT, had insufficient bandwidth to develop and support multiple integrations.
- All relevant applications had API integration capabilities and all of the recommended integrations were possible. However, IT resources were the limiting factor to completing the integrations.
End Result
We delivered a comprehensive assessment report detailing:
- The current architecture and requirements for all relevant business systems
- Barriers to integration which were primarily a lack of IT resources and prioritization to perform the work.
We also outlined the risks of foregoing these integrations. These included:
- Vocabulary Fragmentation: Continued divergence of taxonomies across departments means that systems and data would be less compatible and integration will require extensive mapping and harmonization.
- Inconsistent Content Classification: With no enterprise standard or governance, assets and content will be tagged with siloed labels and taxonomies, reducing discoverability and utility across the organization.
- Accumulating Data Debt: Continually ‘kicking the can down the road’ limits interoperability between systems and any future integration effort will be larger in cost, time, and scale.
Factor’s discovery and assessment process identified a lack of IT resources as the primary barrier to taxonomy integration, and provided a clear path forward. Prioritizing support for system integration and automation would enable harmonization of intranet taxonomies with enterprise standards, enhancing the value of this important resource.
