The Challenge 

The existing approach wasn't enough

This project arose from an imbalance between incidents logged and use of knowledge resources. We saw 5,000 incidents per month but only 1,300 article views, indicating the robust library was not being utilised effectively.

This challenge was significant for several reasons:

 

  • Impact on the Service Desk: High incident volumes meant the service desk was inundated with calls, limiting their ability to focus on other critical tasks such as creating new knowledge content or improving processes.

 

  • Customer SLAs (Service Level Agreement): For our customers, meeting tight SLAs is essential. Reducing incident volumes would help ensure SLA compliance and even improve performance levels.

 

  • Underutilised Knowledge Base: Despite previous efforts, such as promoting top-viewed articles on the IT Service Portal and including knowledge links in ServiceNow email communications, engagement remained low. These methods lacked visibility and immediacy for end users.

 

In short, the existing approaches were not enough. We needed a solution that would put relevant knowledge directly in front of users at the point of need, making self-service the natural first step before logging an incident.

The solution

A human hand and a robotic hand reaching towards each other, fingertips almost touching, against a futuristic digital background with glowing lines and data patterns.

Structural change

As Knowledge Manager, my role was to lead the project from conception to implementation, transforming the idea into a functional solution in our IT Service Portal. This required collaboration with the ServiceNow admin team to confirm feasibility and platform capability.

We then tested and refined using the ServiceNow test instance, simulating real scenarios typing in common phrases like ‘I forgot my password’ to validate article accuracy and relevance, avoiding mismatched results.

 

Implementation followed structured change management with documentation, approvals, and testing. If issues arose, we applied agile testing, retesting until criteria were met. This ensured a robust, user-focused solution aligned with technical and business requirements.

 

While the implementation process was smooth overall, several challenges emerged during the project:

 

  • Metadata Quality: One of the biggest hurdles was inconsistent metadata in 372 knowledge articles, causing irrelevant search results. We analysed real user search terms and updated metadata to match non-technical phrasing, greatly improving article relevance.

 

  • Short Descriptions: Another issue was vague short descriptions. For example, an article might say “BYOD error” without context, making it harder for users to find the right solution. We reviewed and refined these descriptions to make them more specific and user-friendly.

 

  • Ranking and Visibility: We discovered metadata repetition influenced article ranking in search results. By strategically duplicating metadata across related articles, we improved visibility for the most relevant content.

Healthcare Expertise