KDA Program Evolution

Last updated
Save as PDF

The KDA program evolves as the knowledge base and the KCS practices and techniques mature. As our knowledge base approaches critical mass, it becomes valuable to see what we can learn from the patterns of article reuse. To get the KDA program started it takes some investment and a little bit of structure.

The investment is the allocation of a few resources: allocating time for the KDEs to do the analysis. There is also an investment in creating the reports and dashboards to support the KDEs' analysis. Over time, we also want to make data scientist skills and time available to the KDEs to automate some of the analysis tasks.

Early in the KDA program evolution a little bit of structure or guidance for the KDEs helps get them started. There needs to a person who takes the lead in organizing and supporting the KDE activities. In some organizations this may be the KCS Program Manager, or a KDE may take on the role of coordinating. In larger organizations, this may be a part-time responsibility that a manager takes on.

As we start the KDA program, all the KDEs for domains with critical mass are typically doing the same basic analysis techniques for their respective domains. As the KDEs dig into their respective domains, their focus and activities will evolve to become tailored to the needs and opportunities in their domain.

The sequence of the activities in the "Techniques" section of this guide aligns with the typical KDA program evolution. The initial activity is the article reuse analysis or Pareto analysis: reviewing the most frequently used articles. This exercise almost always exposes inconsistency in the knowledge workers' understanding of the importance of accurate linking. Inevitably, a few of the most highly used articles are not resolution articles. That is, they are not relevant (to the request or case), specific, or actionable, and therefore not helpful in identifying opportunities for root cause analysis and corrective action. As we have mentioned, feedback to the knowledge workers about their contribution to, or disruption of, the value of the knowledge base is critical.

The sequence of the summary of KDE activities in the chart below reflects the order in which these activities are implemented.

Activity	Outcome
Analysis of frequently used articles (Pareto analysis) Identify top 20% of articles reused Assess candidates for root cause analysis and corrective action Review findability and accuracy of the most frequently used articles	The first round of analysis will identify a few legitimate pervasive issues that are candidates for root cause analysis and corrective action. The first round of analysis will also often expose opportunities to improve knowledge workers' understanding of the Solve Loop activities (in particular, effective search techniques and frequent accurate linking). Analysis of frequently used articles should be done periodically. The appropriate frequency of this activity depends on the dynamics of the domain: things like volume, speed of change, and the rate of new issues being reported.
Participate on the KCS Council	The Council's understanding of the reuse patterns and pervasive issues within the domain Recognition of the knowledge workers who are doing frequent and accurate linking Identify opportunities to improve the knowledge workers' behaviors and understanding of the Solve Loop and the value of accurate linking Recommendations for focus areas for the coaches
Recommendations for search engine tuning Ideas to improve findability	The analysis of frequently used articles will expose opportunities to improve search
Identify patterns	Identify articles with common symptoms or generic symptoms that have multiple possible resolutions Identify articles that have common causes/resolutions
Design hub articles or resolution paths	Make decisions on the best article structure (hub article or resolution path) to document the issue. Create the article(s) Merge duplicate articles discovered during this process Test the articles, ideally with a representative from the audience the article is intended to serve
Provide article frequency reuse reports to the owners of the products, services, processes or policies with pervasive issues	Provide high-impact resolution paths to Development with root cause analysis Provide high-volume, low-complexity issues to Development
Feedback Management	Identification of knowledge worker and requestor sentiment Recognition for all those who are contributing value to the knowledge base Identification of improvement opportunities
Conduct New vs Known Analysis	Establish new vs known ratio and trend Establish link accuracy and link rate indicators and trends Identify articles that have high internal reuse but are not available through the self-service. Identify articles that are available through self-service but requestors are not finding them and assess findability (usually the article is not in the requestor's context)
Content Gap Analysis	Once the self-service mechanism is up and running, identify search topics that are not successful Identify articles that exist in the KB but are not visible to the requestors Identify issues for which requestors are seeking a resolution but no article exists